docs: migrate Frontend docs to three-tier structure

[dagil-nvidia]

Summary

• Create docs/components/frontend/ with README and guide following the three-tier documentation structure
• Copy openapi.json to docs/reference/api/ for API reference
• Add deprecation notice to docs/frontends/kserve.md pointing to new location
• Add redirect in conf.py from old path to new location
• Update navigation in frontends.rst to include new docs
• Update in-code README to point to new documentation

Test plan

• Docs build passes with no warnings for new files
• Redirect from frontends/kserve works correctly
• Pre-commit hooks pass

Docs build verification

# One-time setup
uv venv .venv-docs
uv pip install --python .venv-docs --group docs

# Build docs
uv run --python .venv-docs --no-project docs/generate_docs.py

Build succeeded with no errors.

Summary by CodeRabbit

• Documentation
  • Reorganized and expanded Frontend documentation with comprehensive setup, configuration, and deployment guides.
  • Created Frontend Overview and Frontend Guide documentation.
  • Marked KServe documentation as deprecated; content migrated to new Frontend Guide with automatic redirect.
  • Added documentation templates to standardize future documentation creation.

[copy-pr-bot]

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

[coderabbitai]

Walkthrough

Reorganizes and expands frontend documentation into docs/components/frontend, adds deprecation notes and redirects for legacy KServe docs, simplifies the in-repo frontend README, and introduces multiple new documentation templates and migration guides.

Changes

Cohort / File(s)	Summary
Frontend docs & README components/src/dynamo/frontend/README.md, docs/components/frontend/README.md, docs/components/frontend/frontend_guide.md	Rewrote component README, added a comprehensive Frontend guide (HTTP/OpenAI-compatible and KServe gRPC integration, examples, config, usage), and removed previous operational/CLI-centric details.
Deprecation, redirects & nav docs/_sections/frontends.rst, docs/conf.py, docs/frontends/kserve.md	Updated toctree to split frontend docs, added redirect from frontends/kserve to new frontend guide, and added deprecation notes to legacy KServe page.
Documentation templates & migration resources docs/templates/README.md, docs/templates/MIGRATION_GUIDE.md, docs/templates/SOURCE_TARGET_MAPPING.md, docs/templates/EXAMPLE_SKILL.md, docs/templates/backend_guide.md, docs/templates/backend_readme.md, docs/templates/component_design.md, docs/templates/component_examples.md, docs/templates/component_guide.md, docs/templates/component_readme.md, docs/templates/feature_backend.md, docs/templates/feature_readme.md, docs/templates/incode_readme.md, docs/templates/infrastructure_readme.md, docs/templates/integration_readme.md	Added ~13+ template and migration files: migration guide, source→target mapping, example migration skill, and many README/guide/design/example templates for components, backends, features, integrations, infrastructure, and in-code stubs.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~25 minutes

Poem

🐇 I hopped through folders, tidy and spry,

moved frontends home where new guides lie.
Templates bloom, old links say goodbye,
Docs dusted clean beneath a bright sky.

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and specifically describes the main change: migrating Frontend documentation to a three-tier structure, which aligns with the primary objective of the changeset.
Description check	✅ Passed	The description covers all required sections from the template: Overview (Summary), Details (bulleted changes), Where to start (implied through Summary), and Related Issues section is present but empty, which is acceptable if no specific issues are being closed.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Actionable comments posted: 2

🤖 Fix all issues with AI agents

In `@docs/components/frontend/frontend_guide.md`:
- Line 142: The table row contains a broken markdown link labeled "[OpenAI API
Reference]" that points to a non-existent OpenAI API doc; locate the markdown
link token "[OpenAI API Reference]" in frontend_guide.md and either create the
missing OpenAI API document named openai.md in the API docs and update the link
target to point to that new file, or remove/replace the link text in the table
and adjust the row so it no longer references the nonexistent file.

In `@docs/components/frontend/README.md`:
- Around line 79-83: The table row linking "[OpenAI API Reference]" currently
points to a non-existent target ("../../api/openai.md"); update that markdown
link in the docs/components/frontend/README.md table so the link target is the
correct existing path (or remove the row if the reference was intentionally
deleted). Locate the table entry with the text "[OpenAI API Reference]" and
replace the href portion "../../api/openai.md" with the actual file path in the
repo (or delete the entire table row) and ensure the markdown link renders
without a 404.

🧹 Nitpick comments (8)

docs/templates/component_design.md (1)

27-31: Consider adding language specifiers to fenced code blocks.

The ASCII diagram (line 27) and pseudocode block (line 53) lack language specifiers, triggering MD040 warnings. While these are intentionally generic, adding text or plaintext would satisfy the linter and align with the CONSISTENCY rule in your own MIGRATION_GUIDE.md: "Code block languages | Always specify".

Suggested fix

-```
+```text
 ┌─────────┐    ┌─────────┐    ┌─────────┐
 │ Input   │───▶│ Process │───▶│ Output  │
 └─────────┘    └─────────┘    └─────────┘

```diff
-```
+```text
 Pseudocode or formula

</details>


Also applies to: 53-55

</blockquote></details>
<details>
<summary>docs/templates/component_examples.md (1)</summary><blockquote>

`27-35`: **Consider clarifying the "DGDR" acronym.**

The placeholder comments mention "minimal DGDR" and "production DGDR" but this acronym may not be immediately clear to all documentation authors using this template. Consider expanding it or adding a brief note.


<details>
<summary>Suggested clarification</summary>

```diff
 ```yaml
-# Add minimal DGDR from existing docs
+# Add minimal DynamoGraph Deployment Resource from existing docs

</details>

</blockquote></details>
<details>
<summary>docs/templates/feature_readme.md (1)</summary><blockquote>

`20-20`: **Consider adding a link placeholder for the Feature Matrix reference.**

Line 20 references "the Feature Matrix" but doesn't provide a link. Adding a placeholder link would help template users remember to add the actual link.


<details>
<summary>Suggested fix</summary>

```diff
-See the Feature Matrix for full compatibility.
+See the [Feature Matrix](../reference/feature-matrix.md) for full compatibility.
+<!-- Update link to actual feature matrix location -->

docs/templates/MIGRATION_GUIDE.md (1)

83-83: Hardcoded line numbers may become stale.

The reference to "lines 40-98" for redirects mapping in docs/conf.py will become outdated as the file evolves (it's already at line 100 in the current version). Consider removing the line numbers or describing it as "the redirects dict" instead.

Suggested fix

-| `docs/conf.py` | Redirects mapping (lines 40-98) |
+| `docs/conf.py` | Redirects mapping (`redirects` dict) |

docs/components/frontend/frontend_guide.md (2)

13-13: Consider hyphenating compound modifier.

The phrase "industry standard protocol" uses a compound modifier before a noun. Style guides recommend hyphenating: "industry-standard protocol".

✍️ Suggested grammar improvement

-[KServe v2 API](https://github.com/kserve/kserve/tree/master/docs/predict-api/v2) is one of the industry standard protocol for machine learning model inference.
+[KServe v2 API](https://github.com/kserve/kserve/tree/master/docs/predict-api/v2) is one of the industry-standard protocols for machine learning model inference.

Note: Also fixes subject-verb agreement ("is one of the protocols" not "is one of the protocol").

---

38-38: Consider hyphenating compound modifiers throughout.

Several instances use compound modifiers before nouns without hyphens (e.g., "tensor based inference", "tensor based messages", "tensor based deployment"). Standard style guides recommend hyphenation for clarity.

Examples:

• Line 38: "tensor based inference" → "tensor-based inference"
• Line 44: "tensor based messages" → "tensor-based messages"
• Line 99: "KServe based backend" → "KServe-based backend"
• Line 103: "tensor based deployment" → "tensor-based deployment"

Also applies to: 44-44, 99-99, 103-103

docs/templates/EXAMPLE_SKILL.md (2)

42-44: Use ATX-style heading for consistency.

The heading on line 42 uses setext style (underlined with dashes). For consistency with the rest of the document, consider using ATX style (# prefix).

📝 Proposed markdown style fix

----
-name: docs-migration
-description: Migrate Dynamo documentation to the 9-category hierarchy. Use when migrating components, backends, features, or other docs to the new structure.
----
+# Skill Metadata
+
+- **Name**: docs-migration
+- **Description**: Migrate Dynamo documentation to the 9-category hierarchy. Use when migrating components, backends, features, or other docs to the new structure.

Alternatively, if the dashed separator is intentional for Cursor skill format, this can be ignored.

---

60-71: Consider adding language labels to fenced code blocks.

The markdown linter flags unlabeled code blocks at lines 60 and 310. While not critical, adding labels improves syntax highlighting and clarity.

📝 Suggested improvements

Line 60 (directory hierarchy):

-```
+```plaintext
 docs/

Line 310 (validation checklist):

-```
+```markdown
 Validation for: [COMPONENT_NAME]

Also applies to: 310-321

[dagil-nvidia]

/ok to test f6daeb7

[dagil-nvidia]

@coderabbitai full review

[coderabbitai]

✅ Actions performed

Full review triggered.

[coderabbitai]

Actionable comments posted: 3

🤖 Fix all issues with AI agents

In `@docs/components/frontend/frontend_guide.md`:
- Around line 13-106: Several compound modifiers in the frontend_guide (e.g.,
"tensor based", "model based", "tensor based deployment", "model metadata as
tensor based deployment", "model config endpoint") should be hyphenated for
clarity (e.g., "tensor-based", "model-based", "tensor-based deployment",
"model-metadata", "model-config endpoint"); also change the verb "supersedes" to
"supersede" in the sentence describing TensorModelConfig::triton_model_config;
update occurrences near references to TensorModelConfig and triton_model_config
and the sections titled "Tensor", "Model Metadata / Config", and "Inference"
accordingly.

In `@docs/frontends/kserve.md`:
- Around line 3-4: Update the note in kserve.md to include a concrete removal
timeline by editing the existing notice text that references "Frontend Guide"
(../components/frontend/frontend_guide.md) — replace the vague "will be removed
in a future release" with a specific version or date (e.g., "will be removed in
vX.Y.Z (expected YYYY-MM-DD)"), so readers can plan migrations.

In `@docs/templates/EXAMPLE_SKILL.md`:
- Around line 60-71: The markdown fenced code blocks in the EXAMPLE_SKILL.md
snippet are missing language identifiers (triggering MD040); update both fenced
blocks that contain the docs tree and the "Validation for: [COMPONENT_NAME]"
checklist to use a language token such as text (i.e., change ``` to ```text) so
linting passes; look for the code fences surrounding the directory tree and the
validation checklist in the file and add the language identifier there.

🧹 Nitpick comments (2)

docs/templates/component_design.md (2)

27-31: Consider specifying language for ASCII diagram block.

While optional for templates, specifying a language (e.g., text) for the ASCII diagram code block would satisfy markdown linting rules and provide clearer guidance for template users.

📝 Optional improvement

-```
+```text
 ┌─────────┐    ┌─────────┐    ┌─────────┐
 │ Input   │───▶│ Process │───▶│ Output  │
 └─────────┘    └─────────┘    └─────────┘

</details>

---

`53-55`: **Consider specifying language for pseudocode block.**

Adding a language identifier (e.g., `text` or `pseudocode`) would align with markdown best practices.



<details>
<summary>📝 Optional improvement</summary>

```diff
-```
+```text
 Pseudocode or formula

</details>

</blockquote></details>

</blockquote></details>

<!-- This is an auto-generated comment by CodeRabbit for review status -->

[dagil-nvidia]

/ok to test 3e615b0

[dagil-nvidia]

/ok to test 6c14efc

[dagil-nvidia]

/ok to test 4958ec2

[dagil-nvidia]

/ok to test 70b39d4

[dagil-nvidia]

/ok to test 3c3953e

[dagil-nvidia]

/ok to test 4fca502

[dagil-nvidia]

/ok to test 4095968