[None][fix] source DSA metadata from sparse params

[liji-nv]

Summary by CodeRabbit

• Bug Fixes

  • Improved sparse attention metadata handling for DeepSeek models.
  • Added a default index head dimension of 128 when no value is provided.
  • Ensured cache geometry and compression settings consistently use sparse metadata values.
  • Added validation for incompatible DeepSeek v4 metadata configurations.

• Tests

  • Added coverage verifying sparse metadata correctly determines indexer dimensions, compression ratios, and cache geometry.

Description

Test Coverage

PR Checklist

Please review the following before submitting your PR:

• PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

• PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

• Test cases are provided for new code paths (see test instructions)

• If PR introduces API changes, an appropriate PR label is added - either api-compatible or api-breaking. For api-breaking, include BREAKING in the PR title.

• Any new dependencies have been scanned for license and vulnerabilities

• CODEOWNERS updated if ownership changes

• Documentation updated as needed

• Update tava architecture diagram if there is a significant design change in PR.

• The reviewers assigned automatically/manually are appropriate for the PR.

• Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

To see a list of available CI bot commands, please comment /bot help.

[liji-nv]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #58658 [ run ] triggered by Bot. Commit: 5f6b79b Link to invocation