Updated docs and added migration guide

[gdcsinaptik]

This PR enhances the PandasAI documentation by adding comprehensive migration guides and troubleshooting resources to help users transition from v2 to v3, along with general documentation improvements.

---

Important

Enhanced PandasAI documentation with migration guides, troubleshooting, and updates for v3 features, including configuration, LLM setup, and semantic layer.

• Migration Guides:
  • Added comprehensive migration guide from v2 to v3 in migration-guide.mdx.
  • Detailed backwards compatibility in migration-backwards-compatibility.mdx.
  • Troubleshooting tips for migration issues in migration-troubleshooting.mdx.
• Documentation Updates:
  • Updated README.md to reflect new installation and usage instructions.
  • Enhanced docs/mint.json for navigation and versioning.
  • Revised docs/v2 and docs/v3 files to align with v3 features, including agent.mdx, skills.mdx, and semantic-layer files.
• Configuration and Setup:
  • Introduced global configuration using pai.config.set().
  • Explained LLM setup with pandasai-litellm in large-language-models.mdx.
  • Detailed new data connector extensions in data-ingestion.mdx.
• Semantic Layer and Transformations:
  • Introduced semantic layer concepts in semantic-layer.mdx.
  • Documented data transformations in transformations.mdx.
  • Explained views and schema creation in views.mdx and new.mdx.

^{This description was created by for 5846f07. You can customize this summary. It will automatically update as commits are pushed.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 91.32%. Comparing base (e342f17) to head (5846f07).
⚠️ Report is 12 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1832      +/-   ##
==========================================
- Coverage   91.33%   91.32%   -0.01%     
==========================================
  Files          71       72       +1     
  Lines        2803     2813      +10     
==========================================
+ Hits         2560     2569       +9     
- Misses        243      244       +1     

Flag	Coverage Δ
unittests	91.32% <ø> (-0.01%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[ellipsis-dev]

Caution

Changes requested ❌

Reviewed everything up to 1a33279 in 2 minutes and 28 seconds. Click for details.

• Reviewed 2963 lines of code in 26 files
• Skipped 0 files when reviewing.
• Skipped posting 9 draft comments. View those below.
• Modify your settings and rules to customize what types of comments Ellipsis leaves. And don't forget to react with 👍 or 👎 to teach Ellipsis.

1. docs/v3/semantic-layer/new.mdx:40

• Draft comment:
  Syntax issue: Missing comma between the 'sale_date' key and its subsequent key-value pair. Ensure to separate dictionary entries with a comma.
• Reason this comment was not posted:
  Comment was not on a location in the diff, so it can't be submitted as a review comment.

2. docs/v3/migration-guide.mdx:87

• Draft comment:
  Model naming inconsistency: Some examples use 'gpt-4.1-mini' while others use 'gpt-4o-mini'. Standardize the model name throughout the docs.
• Reason this comment was not posted:
  Comment looked like it was already resolved.

3. docs/v3/semantic-layer/new.mdx:25

• Draft comment:
  Stylistic note: When passing keyword arguments (e.g. 'df = file'), consider removing extra spaces for consistency (e.g. 'df=file').
• Reason this comment was not posted:
  Confidence changes required: 50% <= threshold 50% None

4. docs/v3/enterprise-features.mdx:12

• Draft comment:
  Content clarity: The Enterprise Features section clearly lists connectors and features with proper styling. No changes required here.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50% None

5. docs/v3/skills.mdx:6

• Draft comment:
  Good use of the '@pai.skill()' decorator and thorough examples; documentation for skills is comprehensive.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50% None

6. docs/mint.json:67

• Draft comment:
  Typo possibly: The URL 'v3/semantic-layer/semantic-layer' appears to have a duplicated segment. Please confirm if this is intentional or if it should be corrected (e.g., 'v3/semantic-layer').
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% This appears to be a documentation structure change where content is being reorganized. The duplicated segment could be intentional for organizational purposes. The comment is asking for confirmation rather than pointing out a clear issue. According to the rules, we should not ask authors to confirm their intentions or make speculative comments. The path structure could be confusing for users navigating the documentation. The duplicate segment might make the URL less elegant. The documentation structure is an intentional organization choice, and there's no clear evidence that this is a mistake rather than a deliberate decision. The rules specifically state not to ask authors to confirm their intentions. The comment should be deleted as it's asking for confirmation of intention rather than pointing out a clear issue that needs to be fixed.

7. docs/v2/connectors.mdx:208

• Draft comment:
  There is a typographical error on this line. "datasests" should be corrected to "datasets".
• Reason this comment was not posted:
  Comment was on unchanged code.

8. docs/v3/getting-started.mdx:72

• Draft comment:
  Typographical note: The sentence "Find it more about output data formats [here]..." seems awkward. Consider revising to something like "Learn more about output data formats [here]" for clarity.
• Reason this comment was not posted:
  Comment was on unchanged code.

9. docs/v3/semantic-layer/data-ingestion.mdx:123

• Draft comment:
  Typographical error: "a enterprise cloud data extension" should be "an enterprise cloud data extension".
• Reason this comment was not posted:
  Comment was on unchanged code.

Workflow ID: wflow_bsJ8cHhCfUH0yEpf

^{You can customize by changing your verbosity settings, reacting with 👍 or 👎, replying to comments, or adding code review rules.}

[ellipsis-dev]

Important

Looks good to me! 👍

Reviewed 5846f07 in 2 minutes and 20 seconds. Click for details.

• Reviewed 214 lines of code in 1 files
• Skipped 0 files when reviewing.
• Skipped posting 5 draft comments. View those below.
• Modify your settings and rules to customize what types of comments Ellipsis leaves. And don't forget to react with 👍 or 👎 to teach Ellipsis.

1. docs/v3/migration-guide.mdx:389

• Draft comment:
  The removal comment here omits 'custom_head', which is included in the removed options list above. Please update for consistency.
• Reason this comment was not posted:
  Comment was on unchanged code.

2. docs/v3/migration-guide.mdx:441

• Draft comment:
  Consider reformatting the 'Next Steps' section as a proper bullet list (one item per line) to improve readability.
• Reason this comment was not posted:
  Comment did not seem useful. Confidence is useful = 0% <= threshold 50% This comment is purely about formatting and does not address any code-related issues or improvements. It doesn't suggest any specific code changes or improvements related to functionality, testing, or best practices.

3. docs/v3/migration-guide.mdx:288

• Draft comment:
  Consider rephrasing 'In this way you can decide if migrating or not.' to 'This way, you can decide whether to migrate.' for clarity.
• Reason this comment was not posted:
  Comment was on unchanged code.

4. docs/v3/migration-guide.mdx:364

• Draft comment:
  The hyperlink text is split across two lines ("Enterprise" on one line and "Features" on the next). This could potentially break markdown rendering of the link. Consider merging the broken link text back onto a single line.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% Looking at the full file context, this appears to be a documentation file using MDX format with React-style Note components. The line wrapping appears to be intentional for readability and the MDX parser should handle this fine. Similar line wrapping is used consistently throughout the file for other links and text within Note components. The comment seems overly cautious about a non-issue. I could be wrong about MDX parsing - maybe some MDX parsers are more strict about link formatting. Also, even if it works, consistent formatting might be desirable. Looking at the rest of the file, this same line-wrapping style is used successfully in multiple places, suggesting it works fine with their toolchain. Making formatting more consistent would be a purely stylistic change. This comment should be deleted as it raises concerns about a non-issue - the line-wrapped markdown link appears to be working as intended within the MDX format.

5. docs/v3/migration-guide.mdx:442

• Draft comment:
  Typo: The markdown link text for 'Backwards Compatibility' is inadvertently split across two lines ("Backwards" on one line and "Compatibility" on the next). Please combine the text into a single line to ensure proper markdown rendering.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% The comment is about formatting in documentation, which could affect readability. However, looking at the full context, this is inside a Note block with bullet points, and the markdown still renders correctly even with the line break. The formatting change is minor and doesn't affect functionality. The split line might make the source markdown slightly harder to read. However, am I being too picky about a minor formatting issue that doesn't affect the rendered output? Since the markdown still renders correctly and this is just about source code readability, this comment is too minor to be worth addressing. Delete this comment as it addresses a minor formatting issue that doesn't affect functionality or rendered output.

Workflow ID: wflow_joJoA3F2cYF5XCkW

^{You can customize by changing your verbosity settings, reacting with 👍 or 👎, replying to comments, or adding code review rules.}