Add differentiable histogram to saxs loss calculation

[stephprince]

Motivation

The current version of the saxs loss calculation was not differentiable and the resulting tensor had no associated gradient function. For the saxs component to contribute to the gradient computation during backprop, I believe a differentiable histogram or equivalent replacement is needed.

I have added two options here with a couple of figures on how they compare to the original function.

How to test the behavior?

python train.py path/to/data path/to/output --use_saxs_loss_only

Checklist

• Did you update CHANGELOG.md with your changes?
• Have you checked our Contributing document?
• Have you ensured the PR clearly describes the problem and the solution?
• Is your contribution compliant with our coding style? This can be checked running ruff from the source directory.
• Have you checked to ensure that there aren't other open Pull Requests for the same change?
• Have you included the relevant issue number using "Fix #XXX" notation where XXX is the issue number? By including "Fix #XXX" you allow GitHub to close issue #XXX when the PR is merged.

[codecov]

Codecov Report

Attention: Patch coverage is 76.27119% with 14 lines in your changes missing coverage. Please review.

Project coverage is 25.02%. Comparing base (9a2a41c) to head (6df9b3f).

Files with missing lines	Patch %	Lines
src/metfish/msa_model/utils/loss.py	57.14%	9 Missing ⚠️
tests/test_extract_seq.py	28.57%	5 Missing ⚠️

❗ There is a different number of reports uploaded between BASE (9a2a41c) and HEAD (6df9b3f). Click for more details.

HEAD has 1 upload less than BASE

Flag	BASE (9a2a41c)	HEAD (6df9b3f)
	2	1

Additional details and impacted files

@@             Coverage Diff             @@
##             main       #7       +/-   ##
===========================================
- Coverage   44.05%   25.02%   -19.04%     
===========================================
  Files           3        6        +3     
  Lines         227     1095      +868     
===========================================
+ Hits          100      274      +174     
- Misses        127      821      +694     

☔ 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.