[Machine Learning] Implement YOLOv8 for Automatic Fish Cropping

[jpdevhub]

Before sending the image to our classification models, use a lightweight object detection model (like YOLOv8-nano) via ONNX Runtime Web to automatically locate and crop the Body, Eye, and Gill from the camera feed.