Add Apple MPS training test

[glenn-jocher]

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Enhanced testing suite for Ultralytics on Apple MPS devices 🍏💻.

📊 Key Changes

• Imported new identifiers (ARM64, MACOS) to improve device-specific checks.
• Added a new test to validate YOLO training specifically on Apple MPS (Metal Performance Shaders) devices.

🎯 Purpose & Impact

• Purpose: To ensure Ultralytics' YOLO models train effectively on devices leveraging Apple's MPS, catering specifically to ARM64 and macOS environments. This inclusivity ensures broader compatibility and performance optimization.
• Impact: Users with Apple MPS devices can expect more reliable performance, with testing now covering these environments. This update promises a smoother experience for a significant user segment utilizing ARM64-based macOS devices for machine learning tasks.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 70.61%. Comparing base (972f615) to head (f75e17a).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #9960      +/-   ##
==========================================
- Coverage   70.61%   70.61%   -0.01%     
==========================================
  Files         122      122              
  Lines       15584    15584              
==========================================
- Hits        11005    11004       -1     
- Misses       4579     4580       +1     

Flag	Coverage Δ
Benchmarks	35.61% <ø> (ø)
GPU	37.37% <ø> (-0.02%)	⬇️
Tests	66.63% <ø> (-0.09%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[glenn-jocher]

Seems like CI runners are preventing access to MPS memory, so test results here are not valid.