SAM Model Predict Bug

I had followed https://github.com/heartexlabs/label-studio-ml-backend/tree/master/label_studio_ml/examples/segment_anything_model to deploy **SAM model**

But there are some bugs of **SAM** predictions:
### The coordinates of the mask will be offset from the target I expected, but in fact the shape of the mask is correct

Examples:
1. Here is a 2048*1024 image performance(**Offset**):

2. Here is a 640*426 image performance(**Correct**):

3. Here is a 640*480 image performance(**Offset**):

There are absolutely bugs on the process after prediction, and I think the key of question is the image padding

Here are the original images:

Start in label_studio_ml/examples/segment_anything_model and trace how SAM predictions are processed after inference, focusing on image padding and coordinate conversion. Reproduce the reported cases using the 2048×1024, 640×426, and 640×480 images; done means the predicted masks align with the target images without offsets.

Written by the indexing model from the issue text.