🚀 Speed up Image Processing for CV models

[yaswanth19]

What does this PR do?

This speeds up image processing by converting the data to numpy array before sending to BatchFeature.

Queries:

• Currently I am checking if all the images are of same size or not before converting the whole batch to ndarray; for that I am utilizing the arguments like do_resize,do_pad etc.. but is it necessary to do ti this way or we can write a helper function which checks whether every sub array is uniform or not and if so then convert the whole array to ndarray. WDYT?

Fixes #31205

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?

@yonigozlan Please review this first draft PR where I have applied the logic only on some model and once approach is finalised, same changes would be made to all other models

[yaswanth19]

@yonigozlan @amyeroberts A gentle ping to address my query so that I can implement it fully

[yaswanth19]

@yonigozlan @amyeroberts A gentle ping again

[yaswanth19]

@amyeroberts @yonigozlan Please do a first review and address my query