Fix affine getter for multipath images

[fepegar]

Fixes #762.

Description
An error was being raised when trying to get the affine of multipath images, as the list of paths was being passed to the SimpleITK reader. Now, the images are loaded before returning the affine if the image is multipath.

Checklist

• I have read the CONTRIBUTING docs and have a developer setup (especially important are pre-commitand pytest)
• Non-breaking change (would not break existing functionality)
• Breaking change (would cause existing functionality to change)
• Tests added or modified to cover the changes
• Integration tests passed locally by running pytest
• In-line docstrings updated
• Documentation updated, tested running make html inside the docs/ folder
• This pull request is ready to be reviewed
• If the PR is ready and there are multiple commits, I have squashed them and force-pushed

@dmus

[codecov]

Codecov Report

Merging #763 (11f3695) into main (f0abd6b) will decrease coverage by 0.00%.
The diff coverage is 95.00%.

@@            Coverage Diff             @@
##             main     #763      +/-   ##
==========================================
- Coverage   93.88%   93.88%   -0.01%     
==========================================
  Files          72       72              
  Lines        4499     4514      +15     
==========================================
+ Hits         4224     4238      +14     
- Misses        275      276       +1     

Impacted Files	Coverage Δ
torchio/data/image.py	87.50% <95.00%> (+0.24%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update f0abd6b...11f3695. Read the comment docs.

[dmus]

Quick fix :) Now you load the image when it is a multipath image. Another option would be to only call read_affine with the first element of the list when it is a multipath? Or is that option not preferable?

[fepegar]

Yes, that's another option. But I thought this would be safer, as it checks that all affines in the input images are the same (I think). I thought that this compromise of speed vs robustness was worth it as passing a sequence of paths is probably a not very common use case. Does that make sense?