Fix post process function called in the instance segmentation example of mask2former

[OnTheThirdDay]

What does this PR do?

• change doc of mask2former example of post_process_instance_segmentation to the right function call.
• change doc of maskformer and mask2former on post_process_instance_segmentation for clarity on working with overlapping instances.
• change examples in doc of segmentation models from image[::-1] to (image.height, image.width) for clarity.

Fixes # (issue)

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

[Rocketknight1]

This looks like a good fix to me, but cc @NielsRogge to confirm!

[NielsRogge]

Thanks for fixing! cc @qubvel

[qubvel]

Thank you!

[qubvel]

It would be also nice to include:

Suggested change

	... outputs, target_sizes=[image.size[::-1]]
	... outputs, target_sizes=[(image.height, image.width)]

[OnTheThirdDay]

Should we change this line in the examples for all three segmentation tasks?

[qubvel]

Yeah, sure!

[OnTheThirdDay]

I just made another change to the doc of post_process_instance_segmentation of both maskformer and mask2former image processors.

I added this line, as current default behavior of the method will not deal with the cases where instances could overlap, but the descriptions related to return_coco_annotation and return_binary_maps, though clearly understandable, are not very intuitive for users to realize they need to use them.

If instances could overlap, set either return_coco_annotation or return_binary_maps to `True` to get the correct segmentation result.

Additionally, I changed this line to make it more precise for the description of the returned result covering the case where return_binary_maps is set to True.

`List[Dict]`: A list of dictionaries, one per image, each dictionary containing two keys:
            - **segmentation** -- A tensor of shape `(height, width)` where each pixel represents a `segment_id`, or
              `List[List]` run-length encoding (RLE) of the segmentation map if return_coco_annotation is set to
              `True`, or a tensor of shape `(num_instances, height, width)` if return_binary_maps is set to `True`.
              Set to `None` if no mask if found above `threshold`.

[OnTheThirdDay]

the pr now has included above mentioned changes on doc.
@NielsRogge @qubvel @ArthurZucker

[ArthurZucker]

🤗 thanks