[trainer] refactor: no need to call load_reward_manager in compute_reward_async

[eric-haibin-lin]

What does this PR do?

Simply make changes in #1406 backward compatible. We'll remove the args for config & tokenizer in next version.
Credit to @emergenz

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

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.

[gemini-code-assist]

Code Review

This pull request refactors compute_reward_async to accept a reward_fn directly, which is a good improvement for decoupling. The changes correctly maintain backward compatibility by allowing the old config and tokenizer arguments, with a deprecation warning.

My review identifies one high-severity issue regarding the use of assert for input validation, which could be problematic if assertions are disabled. I've provided a suggestion to replace it with a more robust ValueError check.