Fix error with invertible custom transforms, also keep include/exclude lists for inverse transforms.

[efirdc]

Description
This fixes two bugs that I found when making a custom invertible Transform.

First is that Subject.history would not find any transforms defined outside torchio. To fix this I added a get_subclasses(target_class) function to utils.py.

Other thing I noticed was that Transform._get_reproducing_arguments() did not keep the include/exclude lists. So I added them there.

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 attempted to squash them and force-push

[codecov]

Codecov Report

Merging #398 (2fe0433) into master (c9d9b42) will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master     #398      +/-   ##
==========================================
+ Coverage   92.41%   92.42%   +0.01%     
==========================================
  Files         117      117              
  Lines        5935     5943       +8     
==========================================
+ Hits         5485     5493       +8     
  Misses        450      450              

Impacted Files	Coverage Δ
torchio/data/subject.py	94.53% <100.00%> (+0.08%)	⬆️
torchio/transforms/transform.py	94.01% <100.00%> (+0.07%)	⬆️
torchio/utils.py	89.25% <100.00%> (+0.36%)	⬆️

---

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 c9d9b42...2fe0433. Read the comment docs.

[fepegar]

Nice. The invertibility feature is very young and I'm sure there is still a lot of room for improvement. Many thanks for your contribution, @efirdc!

[fepegar]

@all-contributors please add @efirdc for code

[allcontributors]

@fepegar

I've put up a pull request to add @efirdc! 🎉

[romainVala]

Hello
just out of curiosity, @efirdc do you use transform outside torchio (which one, how it is compatible with torchio) ?

[efirdc]

Hi @romainVala,

At the time it was the label transformations, but they are included in torchio now. At the moment I'm not using any custom transformations.