chore: map compute_cap from GPU name

[haixiw]

What does this PR do?

Fix the issue that compute_cap is not avaliable from nvidia-smi --query-gpu=compute_cap --format=csv | sed -n '2p' | sed 's/\.//g' when the CUDA driver version is old.

1. add cuda compate script to dynamically switch the use of the CUDA Compatibility Package based on the detected Nvidia driver version on the deployed host . https://docs.aws.amazon.com/sagemaker/latest/dg/inference-gpu-drivers.html#collapsible-cuda-compat
2. create a mapping from GPU name to compute_cap based on the current avaliable instances from AWS hosting.

Fixes # (issue) huggingface/candle#733

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

[OlivierDehaene]

Thanks a lot for this fix!