feat(wrapper): add gpu support in containers

[Fastiraz]

Description

Add GPU support using the torch Python module and the device_requests (--gpus in CLI) Docker argument.
To do this I implemented the isGPUAvailable function which checks for GPU availability and returns the appropriate value for Docker's --gpus argument. Finally, I added docker_args["device_requests"] to enable GPU support when creating a container.

Point of attention

I've also imported the numpy Python module but it's not required.
I imported it because of a warning message that appears when creating a new container with Exegol.

No module named numpy...

It works fine even without numpy, the error is just displayed when the module is missing.
For now, I haven't figured out how to solve this properly.

I found a way to suppress the warning using warning filters, so it doesn't print the message. Let me know if you want me to implement it this way.

Here's an example:

import warnings
warnings.filterwarnings("ignore", message="No module named numpy", category=ImportWarning)

---

[skyf0l]

Can you check for GPU without a installing a 1Go torch dependency ?

[Fastiraz]

Yes but only with some hacks.

Instead of checking properly with a Python module, I can just execute nvidia-smi on a subprocess and check if there is any standard output or standard error.
If there is a standard output, that mean NVIDIA drivers are installed.
Otherwise, if there is a standard error, that mean NVIDIA drivers are not installed.

For macOS, I just have to check to platform and the CPU architecture.
If the platform is macOS and the CPU architecture is ARM, we determine that the "GPU" is MPS.

I've already a version like that and it works well on Linux.
Keep in mind that I cannot test it with AMD GPUs and Linux or Windows operating systems.

If you think a version like this is better, I can make a new commit with the new function.

Here's the new function:

def isGPUAvailable(self) -> Optional[str]:
      import platform, subprocess
      from typing import Optional
      system = platform.system().lower()

      try:
          gpu_names = subprocess.run(
              ['nvidia-smi'],
              stdout=subprocess.PIPE,
              stderr=subprocess.DEVNULL,
              check=True
          ).stdout.decode().strip().split('\n')
          return '"device=0"' if len(gpu_names) == 1 else '"all"'
      except (subprocess.CalledProcessError, FileNotFoundError):
          pass

      if system == "darwin" and platform.machine().startswith("arm"):
          return None

      return None

[mvthul]

Oehhh i would like it!!!

[Macbucheron1]

I am also interested in the merge of this PR

[romainpanno]

Could be usefull for some gpu calculations, hashcat for example.
I'm also interested in the merge of this PR !