[model] chore: deprecated legacy code for GRM

[yyDing1]

What does this PR do?

Deprecate legacy code for GRM as it's now supported in reward loop.

Future PR may further deprecate {fsdp/megatron}_reward_model by the following steps:

1. support discriminative reward models in reward loop
2. make the reward model & router arch stable for large-scale training
3. refactor the config of reward models (in replace with the ones in recipe/fapo/config/rm_config.yaml)
4. deprecate legacy {fsdp/megatron}_reward_models

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 deprecates and removes legacy code for the sglang-based Generative Reward Model (GRM), which is a positive step towards simplifying the codebase. The changes primarily involve deleting the RewardModelWorker and its related engine implementation. While the removal seems largely self-contained, I've identified a potential critical issue regarding a patch for the sglang library that is also being removed. This could affect other parts of the system that rely on sglang.