Add lazy-loading for extras packages

[sjmonson]

TODO

• Pending upstream change: Support lazy importing submodules in attach  scientific-python/lazy-loader#168
• Support for soft extras (E.g. perf group)
• Lazy load in __main__.py to improve CLI responsiveness

Summary

Lazy loads extras submodules in order to defer import errors to the time of use.

Details

TODO

Test Plan

Without vLLM:

1. Run guidellm benchmark run --help and observe no errors
2. Run uv run guidellm benchmark run --backend vllm_python --model test and observe error with helpful message

With vLLM:

1. Run guidellm benchmark run --help and observe no errors
2. Run uv run guidellm benchmark run --backend vllm_python --model test ... and observe successful benchmark
3. Run tox -re test-e2e and observe that tests pass (Previously they would fail when vLLM was installed due to load times)

Related Issues

• Replaces Add helper to defer ImportError on extras and support lazy loading #636

---

• "I certify that all code in this PR is my own, except as noted below."

Use of AI

• Includes AI-assisted code completion
• Includes code generated by an AI application
• Includes AI-generated tests (NOTE: AI written tests should have a docstring that includes ## WRITTEN BY AI ##)

[mergify]

This pull request has merge conflicts that must be resolved before it can be
merged. Please rebase the PR, @sjmonson.

https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork