[trainer] fix: Add data.seed to config

[HollowMan6]

What does this PR do?

Add concise overview of what this PR aims to achieve or accomplish. Reference related GitHub issues and PRs that help with the review.

Resolves #2234

Also support not setting the seed value (null)

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 introduces a data.seed configuration option to ensure reproducible data shuffling, which is a good improvement for experiment consistency. The changes are applied across various configuration files and documentation. However, there are a couple of important issues. First, two auto-generated configuration files have been modified manually, which violates the repository's process. These should be updated by running the generation script after changing the source configs. Second, the implementation does not handle the case where data.seed is set to null, which will lead to a runtime crash. The documentation should be updated to warn users about this, and ideally, the code should be made more robust.

[gemini-code-assist]

Code Review

This pull request introduces a data.seed configuration option to control the seed for data shuffling, which is a useful feature for reproducibility. The implementation correctly handles cases where the seed is provided as an integer or is explicitly set to null (or omitted), allowing for both deterministic and non-deterministic shuffling. However, there is a critical issue in the documentation that needs to be addressed.

[gemini-code-assist]

Code Review

This pull request successfully introduces the data.seed configuration option to enable reproducible data shuffling, which is a valuable addition for experiment consistency. The changes are applied consistently across various configuration files, documentation, and trainer implementations. My review includes a few suggestions to enhance robustness by adding type validation for the new seed parameter, which will help prevent runtime errors from invalid configurations.