Replace pynvml with nvidia-ml-py in requirements

[myhloli]

/lib/python3.12/site-packages/torch/cuda/__init__.py:61: FutureWarning: The pynvml package is deprecated. Please install nvidia-ml-py instead. If you did not install pynvml directly, please report this to the maintainers of the package that installed pynvml for you.
  import pynvml  # type: ignore[import]

[lvhan028]

Hi, @myhloli
Thank you for bringing this to our attention.
You can go ahead and remove pynvml from runtime_cuda.txt as LMDeploy no longer relies on it.
Additionally, would you be willing to remove it from the unit-test.yml file as well? We'd really appreciate your help with that.

[myhloli]

Okay, after I directly uninstalled pynvml and nvidia-ml-py locally, the program runs normally without any exceptions.