fix qwen2 cache_position for PyTorch Engine when transformers>4.41.2

[zhyncs]

Motivation

fix #1885

Hi @lvhan028 @grimoire @RunningLeon @AllentDan May you help review this pr? Thanks.

Modification

as titled

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.

[zhyncs]

ref #1852

[Volta-lemon]

I had this problem and I saw you solved it. Now do I just need to re-execute pip install lmdeploy[all]==0.4.2?

[zhyncs]

@Volta-lemon This PR has been merged and I manually triggered the compilation at https://github.com/zhyncs/lmdeploy-build/actions/runs/9737958121 and https://github.com/zhyncs/lmdeploy-build/actions/runs/9737959536. After the compilation is completed, it can be installed directly.

[zhyncs]

@Volta-lemon https://github.com/zhyncs/lmdeploy-build/releases/tag/1f5dd4e
You may add extra args when installing with pip3 using --force-reinstall --no-deps

[Volta-lemon]

Thank you for solving my problem