llm: register sdpa variant

[lanluo-nvidia]

Description

Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required for this change.

Fixes # (issue)

Type of change

Please delete options that are not relevant and/or add your own.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Checklist:

• My code follows the style guidelines of this project (You can use the linters)
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas and hacks
• I have made corresponding changes to the documentation
• I have added tests to verify my fix or my feature
• New and existing unit tests pass locally with my changes
• I have added the relevant labels to my PR in so that relevant reviewers are notified

[peri044]

Minor comments. Please update the supported models section in README.md and docs/user_guide.

[peri044]

1. Could you quickly try running these variants as well ?
   a) google/gemma-3-4b-it
   b) google/gemma-3-270m-it
2. Please update the supported model list here as well: https://github.com/pytorch/TensorRT/blob/main/docsrc/tutorials/compile_hf_models.rst
3. Could you add a testcase for 1 Gemma-3 decoder layer with sliding window attention ? The test case could be located at https://github.com/pytorch/TensorRT/tree/main/tests/py/dynamo/models as test_llm_models.py

[peri044]

LGTM