[LMM] Implement merged multimodal processor for whisper

[Isotr0py]

TODO

• Fix profiling issue
• Add processor test

[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.

🚀

[DarkLight1337]

cc @ywang96 is it safe to change the encoder prompt given the change in vllm/multimodal/profiling.py?

[DarkLight1337]

LGTM. Just need @ywang96 to comment on the profiling changes.

[DarkLight1337]

After this PR, can you update the model development docs with some info on how to implement merged multi-modal processor for encoder-decoder models?

[Isotr0py]

can you update the model development docs with some info on how to implement merged multi-modal processor for encoder-decoder models?

Yes, I will add a doc using whisper and florence-2 as e2e example respectively after this PR and #13320 merged, because there is also a minor change on EncDecMultimodalProcessor to fit text-only x-attn VLM in that PR. (That PR should be ready before the end of this weekend)