Add new checkpoints for training
Add yolov9*-seg (image segmentation) models and yolov10 models (object detection).
We might have to refactor the way we handle these checkpoints (since yolov**.pt
does not always mean Image classification anymore).