[recipe, fsdp] feat: support GPT-OSS-20B DAPO training script on ASCEND NPU

[mikequan0425]

What does this PR do?

Provide an script for DAPO-training GPT-OSS-20B on NPU

Checklist Before Starting

• Search for similar PRs. Paste at least one query link here: [recipe] feat: support qwen2.5-32B DAPO training script on ASCEND NPU #3146, [megatron] feat: support gpt-oss #4323
• 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, cfg, reward
  • 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).)

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[gemini-code-assist]

Code Review

This pull request adds a new training script for DAPO with GPT-OSS-20B on Ascend NPUs, along with a documentation update. The changes look good overall. I've found one potential high-severity issue in the new training script where an incorrect advantage estimator might be configured, which could affect the correctness of the training process. My detailed feedback is in the review comment.

[gemini-code-assist]

The script is configured for a DAPO training run, as indicated by the script name, configuration files, and reward manager. However, the advantage estimator is set to grpo. This appears to be inconsistent and likely incorrect for a DAPO recipe. Using a grpo estimator may not align with the DAPO algorithm, potentially leading to incorrect training behavior. It should be changed to the appropriate estimator for DAPO, which is presumably dapo.

Suggested change

	adv_estimator=grpo
	adv_estimator=dapo