Fix mla with spec tokens

[RunningLeon]

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily receiving feedbacks. If you do not understand some items, don't worry, just make the pull request and seek help from maintainers.

Motivation

Fix mla with spec tokens

Modification

Please briefly describe what modification is made in this PR.

BC-breaking (Optional)

Does the modification introduce changes that break the backward-compatibility of the downstream repositories?
If so, please describe how it breaks the compatibility and how the downstream projects should modify their code to keep compatibility with this PR.

Use cases (Optional)

If this PR introduces a new feature, it is better to list some use cases here, and update the documentation.

Checklist

1. Pre-commit or other linting tools are used to fix the potential lint issues.
2. The modification is covered by complete unit tests. If not, please add more unit tests to ensure the correctness.
3. If the modification has a dependency on downstream projects of a newer version, this PR should be tested with all supported versions of downstream projects.
4. The documentation has been modified accordingly, like docstring or example tutorials.

[Copilot]

Pull request overview

This PR fixes a bug in the MLA (Multi-head Latent Attention) implementation to support speculative decoding with multiple tokens. The issue occurs when max_q_seqlen > 1, where the original squeeze(1) operation would fail because the dimension being squeezed is not 1.

Changes:

• Replace squeeze(1) with flatten(0, 1) in the flash_mla_decoding function to properly merge batch and sequence dimensions regardless of sequence length

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.