Consider adding remove_small_objects for SC seg

[valosekj]

I noticed that for some images, the SCIseg model segments "several components" when running spinal cord segmentation inference:

We could consider adding the remove_small_objects function to postprocessing to filter such components.

Note: I have noticed this only for a single subject so far.

[naga-karthik]

I think this feature is now supported with SCT? using the following help:

PARAMETERS:
  -thr <float>          Binarize segmentation with specified threshold. Set to 0 for no thresholding (i.e., soft segmentation). Default value is model-specific and was set during
                        optimization (more info at https://github.com/sct-pipeline/deepseg-threshold).
  -largest KEEP_LARGEST Keep the largest connected-objects from the output segmentation. Specify the number of objects to keep.To keep all objects, set to 0
  -remove-small REMOVE_SMALL [REMOVE_SMALL ...]
                        Minimal object size to keep with unit (mm3 or vox). A single value can be provided or one value per prediction class. Single value example: 1mm3, 5vox. Multiple
                        values example: 10 20 10vox (remove objects smaller than 10 voxels for class 1 and 3, and smaller than 20 voxels for class 2).