[docker] feat: update to vllm 0.10.0, mcore 0.13, transformers 4.55.4

[ETOgaosion]

What does this PR do?

[docker] feat: update to vllm 0.10.0, mcore 0.13, transformers 4.55.4

Checklist Before Starting

• Search for similar PRs. Paste at least one query link here: ...
• Format the PR title as [{modules}] {type}: {description} (This will be checked by the CI)
  • {modules} include fsdp, megatron, sglang, vllm, rollout, trainer, ci, training_utils, recipe, hardware, deployment, ray, worker, single_controller, misc, perf, model, algo, env, tool, ckpt, doc, data
  • If this PR involves multiple modules, separate them with , like [megatron, fsdp, doc]
  • {type} is in feat, fix, refactor, chore, test
  • If this PR breaks any API (CLI arguments, config, function signature, etc.), add [BREAKING] to the beginning of the title.
  • Example: [BREAKING][fsdp, megatron] feat: dynamic batching

Test

For changes that can not be tested by CI (e.g., algorithm implementation, new model support), validate by experiment(s) and show results like training curve plots, evaluation results, etc.

API and Usage Example

Demonstrate how the API changes if any, and provide usage example(s) if possible.

# Add code snippet or script demonstrating how to use this

Design & Code Changes

Demonstrate the high-level design if this PR is complex, and list the specific changes.

Checklist Before Submitting

Important

Please check all the following items before requesting a review, otherwise the reviewer might deprioritize this PR for review.

• Read the Contribute Guide.
• Apply pre-commit checks: pre-commit install && pre-commit run --all-files --show-diff-on-failure --color=always
• Add / Update the documentation.
• Add unit or end-to-end test(s) to the CI workflow to cover all the code. If not feasible, explain why: ...
• Once your PR is ready for CI, send a message in the ci-request channel in the verl Slack workspace. (If not accessible, please try the Feishu group (飞书群).)

[gemini-code-assist]

Code Review

This pull request updates several key dependencies, including vLLM to 0.10.0, Megatron-Core to 0.13, and Transformers to 4.55.4. The changes primarily involve updating version numbers in Dockerfiles and documentation, and adapting the code to the new APIs of these libraries.

My review found one high-severity issue in the monkey-patching logic for vision models. The current patch for transformers>=4.55.0 is likely ineffective, necessitating a performance-degrading workaround that disables flash attention for vision models when using Ulysses sequence parallelism. I've suggested a more robust patching strategy that should fix the underlying issue and allow for the removal of the workaround.

Other changes appear to be correct and consistent with the goal of updating dependencies.

[vermouth1992]

Please modify the name to _ulysses_flash_attention_forward_transformers_4_55

[ETOgaosion]

@vermouth1992 I fix vllm as sglang's way of load_format setting.