New ONNX, TorchScript, CoreML export test matrices

[glenn-jocher]

🛠️ PR Summary

_{Made with ❤️ by Ultralytics Actions}

🌟 Summary

Enhancements in YOLO model export functionality across formats including TorchScript, ONNX, OpenVINO, CoreML, TFLite, *.pb, Paddle, and NCNN.

📊 Key Changes

• Simplified test function descriptions for clarity.
• Restructured export parameter lists for better readability and maintainability.
• Added comprehensive testing across multiple export formats with varying conditions (e.g., batch size, dynamic).
• Implemented CoreML export assertion to ensure the batch size is set to 1, enhancing export compatibility.

🎯 Purpose & Impact

• Clarity: The simplified descriptions and restructured test definitions make it easier for developers to understand and modify export tests.
• Coverage: Extensive testing ensures robustness and reliability of exports across a wide variety of formats and conditions, which is crucial for developers needing to deploy models in different environments.
• Compatibility: The CoreML export now explicitly checks for an unsupported batch size scenario, preventing runtime errors and streamlining the model export process on macOS and Linux.
• User Experience: These changes collectively aim to improve the developer's experience by making the model export process more predictable and error-resistant, benefiting users who deploy YOLO models in diverse environments. 🚀

[codecov[bot]]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 70.69%. Comparing base (2583f84) to head (9ba2ab4).

:exclamation: Current head 9ba2ab4 differs from pull request most recent head 57a01df. Consider uploading reports for the commit 57a01df to get more accurate results

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #11652 +/- ## ========================================== + Coverage 70.60% 70.69% +0.08% ========================================== Files 122 122 Lines 15583 15584 +1 ========================================== + Hits 11003 11017 +14 + Misses 4580 4567 -13 ``` | [Flag](https://app.codecov.io/gh/ultralytics/ultralytics/pull/11652/flags?src=pr&el=flags&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | Coverage Δ | | |---|---|---| | [Benchmarks](https://app.codecov.io/gh/ultralytics/ultralytics/pull/11652/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | `35.67% <100.00%> (+0.06%)` | :arrow_up: | | [GPU](https://app.codecov.io/gh/ultralytics/ultralytics/pull/11652/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | `37.37% <0.00%> (-0.01%)` | :arrow_down: | | [Tests](https://app.codecov.io/gh/ultralytics/ultralytics/pull/11652/flags?src=pr&el=flag&utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics) | `66.78% <100.00%> (+0.06%)` | :arrow_up: | Flags with carried forward coverage won't be shown. [Click here](https://docs.codecov.io/docs/carryforward-flags?utm_medium=referral&utm_source=github&utm_content=comment&utm_campaign=pr+comments&utm_term=ultralytics#carryforward-flags-in-the-pull-request-comment) to find out more.

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[glenn-jocher]

@Burhan-Q trying this out now that it's working well for OpenVINO. ONNX is trickier though, FP16 export is only supported on CUDA as far as I know, and we lack INT8 support, so the matrix is a little less fun.

Torchscript even more so.