Multi shape support

[JairMathAI]

I add some helper functions to train models for a regression task with volumes of variable size , the models who support now regression task are AttentionUnet, Probabilistic Unet and I include the implementation of the nnUnet framework that support regression and in the case of segmentation it include the support of the topological regularizer terms.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 25.76687% with 605 lines in your changes missing coverage. Please review.
✅ Project coverage is 16.49%. Comparing base (ba85ae2) to head (68c9fd5).
⚠️ Report is 34 commits behind head on main.

Files with missing lines	Patch %	Lines
mmv_im2im/utils/custom_transforms.py	0.00%	131 Missing ⚠️
mmv_im2im/proj_trainer_multishape.py	0.00%	90 Missing ⚠️
mmv_im2im/proj_tester.py	8.33%	77 Missing ⚠️
mmv_im2im/utils/variable_collate.py	0.00%	62 Missing ⚠️
mmv_im2im/models/pl_FCN.py	0.00%	42 Missing ⚠️
mmv_im2im/models/pl_nnUnet.py	0.00%	36 Missing ⚠️
mmv_im2im/utils/variable_datamodule.py	0.00%	35 Missing ⚠️
mmv_im2im/models/pl_ProbUnet.py	0.00%	28 Missing ⚠️
mmv_im2im/utils/gdl_regularized.py	26.08%	16 Missing and 1 partial ⚠️
mmv_im2im/utils/elbo_loss.py	36.00%	9 Missing and 7 partials ⚠️
... and 8 more

Additional details and impacted files

@@            Coverage Diff             @@
##            main     #121       +/-   ##
==========================================
+ Coverage   4.40%   16.49%   +12.08%     
==========================================
  Files         36       57       +21     
  Lines       2815     6075     +3260     
  Branches       0     1001     +1001     
==========================================
+ Hits         124     1002      +878     
- Misses      2691     4922     +2231     
- Partials       0      151      +151     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.