support cache_seqlen on recurrent-gdr and causal-conv1d-update

[grimoire]

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

Please describe the motivation of this PR and the goal you want to achieve through this PR.

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 extends the CUDA TileLang kernels to support cache_seqlens-based circular-buffer semantics for (1) recurrent gated-delta-rule state updates and (2) causal_conv1d_update, and expands the test suite to exercise non-trivial sequence lengths and cache behavior.

Changes:

• Add cache_seqlens / circular-buffer support to fused_recurrent_gated_delta_rule (kernel + Python wrapper).
• Add cache_seqlens / circular-buffer support to causal_conv1d_update (kernel + Python wrapper).
• Expand/parametrize kernel tests to cover seqlen > 1, optional cache_seqlens, and optional state indices.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 9 comments.

File	Description
lmdeploy/pytorch/kernels/cuda/gated_delta_rule.py	Extends recurrent GDR kernel/wrapper to handle multi-step seqlen and circular-buffer state addressing via cache_seqlens.
lmdeploy/pytorch/kernels/cuda/causal_conv1d.py	Extends causal_conv1d_update kernel/wrapper to optionally treat conv_state as a circular buffer addressed by cache_seqlens.
tests/pytorch/kernel/test_gated_delta_rule.py	Adds a new circular-buffer correctness test and expands seqlen parameterization.
tests/pytorch/kernel/test_causal_conv1d.py	Adjusts batch size and adds parameterization to test cache_seqlens and optional conv_state_indices.

---

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

You can also share your feedback on Copilot code review. Take the survey.