[Model] Add LoRA support for TransformersModel

[jeejeelee]

• Add unit test
• Update docs

[github-actions]

👋 Hi! Thank you for contributing to the vLLM project.

💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels.

Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which starts running only a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of those by going to your fastcheck build on Buildkite UI (linked in the PR checks section) and unblock them. If you do not have permission to unblock, ping simon-mo or khluu to add you in our Buildkite org.

Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging.

To run CI, PR reviewers can either: Add ready label to the PR or enable auto-merge.

🚀

[hmellor]

This is nice, thank you for adding LoRA support!

I've made a suggestion that allows the can_replace_layer methods to get the parent_cls of the HFCompatibleLinear so that the equivalent of _is_hf_linear can live inside HFCompatibleLinear. Let me know what you think!

[hmellor]

Thanks for making the suggested changes.

One final suggestion is that we reuse the subclassing used for HFCompatibleLinear to create a HFCompatibleLoRA. It feels like less of a hack than replacing the forward method of the instance itself.

[hmellor]

LGTM!