Add perturbation signature calculation from replicate control cells

[Lilly-May]

PR Checklist

• If you've fixed a bug or added code that should be tested, add tests!
• Documentation in docs is updated

Description of changes

As discussed, we would like to have an alternative to the standard Mixscape perturbation signature calculation. Instead of using the X nearest control cells as a reference, this approach considers all control cells within the same replicate. This change (and the corresponding test) is introduced in this PR. Specifically, I added the parameter ref_selection_mode to the perturbation_signature method, allowing users to select which mode to use: nn for the standard approach or manual for selection via the specified replicate method (I’m open to suggestions for better names!).

I also added the preprint citation to the docs, since I noticed that was missing.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 69.56522% with 14 lines in your changes missing coverage. Please review.

Project coverage is 64.92%. Comparing base (218ccb3) to head (0087dcd).

Files with missing lines	Patch %	Lines
pertpy/tools/_mixscape.py	69.56%	14 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #695      +/-   ##
==========================================
+ Coverage   64.89%   64.92%   +0.02%     
==========================================
  Files          46       46              
  Lines        5994     6004      +10     
==========================================
+ Hits         3890     3898       +8     
- Misses       2104     2106       +2     

Files with missing lines	Coverage Δ
pertpy/tools/_mixscape.py	80.27% <69.56%> (-0.02%)	⬇️

[Zethson]

Thank youuuu!