Update CoreML exports to support newer *.mlpackage outputs (#13222)

* Implement and default mlpackage generation for CoreML model exports

Signed-off-by: Ryan Hirasaki <ryanhirasaki@gmail.com>

* Provide command line argument to export as *.mlmodel instead of *.mlpackage for CoreML

Signed-off-by: Ryan Hirasaki <ryanhirasaki@gmail.com>

* Remove macOS check for CoreML quantization

Requirements for macOS during quantization was removed from coremltools 6.0

Signed-off-by: Ryan Hirasaki <ryanhirasaki@gmail.com>

* Undo removal of warning catching

Signed-off-by: Ryan Hirasaki <ryanhirasaki@gmail.com>

* Change file extension references from mlmodel to mlpackage

Signed-off-by: Ryan Hirasaki <ryanhirasaki@gmail.com>

* Auto-format by https://ultralytics.com/actions

---------

Signed-off-by: Ryan Hirasaki <ryanhirasaki@gmail.com>
Co-authored-by: UltralyticsAssistant <web@ultralytics.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: Ultralytics Assistant <135830346+UltralyticsAssistant@users.noreply.github.com>

detect.py

@@ -20,7 +20,7 @@ Usage - formats:
                                  yolov5s.onnx               # ONNX Runtime or OpenCV DNN with --dnn
                                  yolov5s_openvino_model     # OpenVINO
                                  yolov5s.engine             # TensorRT
-                                 yolov5s.mlmodel            # CoreML (macOS-only)
+                                 yolov5s.mlpackage          # CoreML (macOS-only)
                                  yolov5s_saved_model        # TensorFlow SavedModel
                                  yolov5s.pb                 # TensorFlow GraphDef
                                  yolov5s.tflite             # TensorFlow Lite