[Migration] Sendnn inference rename

[joerunde]

Description

This PR updates the vllm_spyre package to sendnn_inference.

Docs and publication workflows have been updated in preparation to publish this as sendnn-inference==2.0.0

❗❗❗

Breaking configuration changes!

• The plugin name for VLLM_PLUGINS is now sendnn_inference
• All config options are now SENDNN_INFERENCE_*
• The precompiled model parser now expects sendnn_inference_version

Related Issues

Closes torch-spyre/spyre-inference#8

Test Plan

• CPU unit tests
• Spyre unit tests
• Spot checks with full model integration tests
• Ensure Test PyPI deploy works

Checklist

• I have read the contributing guidelines
• My code follows the project's code style (run bash format.sh)
• I have added tests for my changes (if applicable)
• I have updated the documentation (if applicable)
• My commits include a Signed-off-by: line (DCO compliance)

[github-actions]

👋 Hi! Thank you for contributing.
Just a reminder: Make sure that your code passes all the linting checks, otherwise your PR won't be able to be merged. To do so, run ./format.sh.
Now you are good to go 🚀.

We also recommend installing prek and configuring it to check your code before every local commit.

[joerunde]

If we want to keep pushing docker images (that nobody uses?) then we'll also need to update our quay configs. I've set things up so that they could run, but disabled the build for now.

PyPI and Test PyPI projects have been updates to publish the new sendnn-inference project once this PR merges

[joerunde]

bot:test
MARKERS="spyre and prefix_caching and not quantized and not multi"

[joerunde]

bot:test
MARKERS="spyre and chunked_prefill"

[Nhan-Hoang]

@joerunde yes I think it is overlapping. We can close this as dup. I actually had this ready last week just for the documentation but fail to submit.

[sducouedic]

I think these two changes are still missing:

• _local_envs_for_test.sh line 10 VLLM_SPYRE_DYNAMO_BACKEND
• .yapfignore: line 1 vllm_spyre/model_executor/...

[rafvasq]

Assuming docker/Dockerfile.amd64 is being left alone for now, a few other references to "vllm spyre" include:

• CONTRIBUTING.md
• AFTU tests docstrings (graph_compare_utils.py, test_compare_graphs.py)
• e2e/test_spyre_mm.py (here)

[tjohnson31415]

Adding some of my searching as well, with commands (overlaps with above comments except for some more changes in .github):

# General search (excluding dot files/dirs)
# excluding docs since that intends to use old names in some places
$ grep -ri 'vllm.spyre' * | grep -v -e '.egg-info' -e '__pycache__' -e 'Binary file' -e 'docs/' -e 'docker/'
_local_envs_for_test.sh:export VLLM_SPYRE_DYNAMO_BACKEND=eager
CONTRIBUTING.md:# Contributing to vLLM Spyre
CONTRIBUTING.md:For details on contributing to vLLM-Spyre, see the **[contributing guide](https://docs.vllm.ai/projects/spyre/en/latest/contributing)**.
examples/online_inference/spyre_vllm_benchmark.py:    parser = argparse.ArgumentParser(description="VLLM Spyre inference benchmarking script.")
tests/aftu/test_compare_graphs.py:"""Compare graphs generated by vLLM-Spyre vs AFTU.
tests/aftu/test_compare_graphs.py:This test compares computation graphs generated by vLLM-Spyre against those
tests/aftu/graph_compare_utils.py:"""Utilities for comparing computation graphs between vLLM-Spyre and AFTU.
tests/e2e/test_spyre_mm.py:    # and vllm spyre running with the eager backend.

# Search dotfiles
$ grep -d skip -i 'vllm.spyre' .*
.yapfignore:vllm_spyre/model_executor/model_loader/spyre_setup.py

# Recurse into specific . directories (avoid .git, .venv, etc)
$ grep -ir 'vllm.spyre' .github/ .vscode/
.github/CODEOWNERS:# This lists cover the "core" components of vLLM-Spyre that require careful review
.github/ISSUE_TEMPLATE/config.yml:    about: Read the vLLM Spyre documentation
.github/ISSUE_TEMPLATE/config.yml:    about: Learn how to contribute to vLLM Spyre
.github/ISSUE_TEMPLATE/feature-request.yml:      Why is this feature important? How would it benefit vLLM Spyre users?

[joerunde]

Travis' commands are now clean- y'all think this is good to go?

[sducouedic]

lgtm!

[joerunde]

good luck us!

[joerunde]

test pypi deployment was successful! 🎉