[veomni] feat: bump veomni to v0.1.8

[deerlu]

What does this PR do?

Bump veomni to v0.1.8

• Fix parallel_state init parameter (ep_size -> extra_parallel_sizes) and use set type for basic_modules
• Automatically rewrites flash_attention_2/3/4 to VeOmni SP-aware variants
• Add _prepare_veomni_flash_attention_kwargs to precompute cu_seq_lens for packed sequences
• Slice position_ids when sp_enabled

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

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, veomni, 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, fully_async, one_step_off
  • 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).)
• If your PR is related to the recipe submodule, please also update the reference to the submodule commit via git submodule update --remote or cd recipe && git pull origin main.

[gemini-code-assist]

Code Review

This pull request introduces support for Flash Attention in the VeOmni engine by adding a utility to derive sequence length metadata from packed position IDs. It also refactors the parallel configuration to use a tuple for extra parallel sizes and improves the deduplication of basic modules for FSDP. Feedback was provided to ensure that the generated sequence length tensors are moved to the same device as the input position IDs to avoid potential device mismatch errors.

[wuxibin89]

@deerlu Do use need to upgrade veomni in ci? It's still use ByteDance-Seed/VeOmni.git@v0.1.4

[deerlu]

@deerlu Do use need to upgrade veomni in ci? It's still use ByteDance-Seed/VeOmni.git@v0.1.4

done, bumped to v0.1.8 (released today)