Add cpa

[szalata]

Describe your changes

This code adds the cpa method. It is a major change since the method operates on single-cell data and outputs an h5ad that contains single-cell data and will have to be pseudobulked and run through limma to get the final output. I limited this PR to CPA, but I'd appreciate your ideas on how to best integrate it into run_benchmark.

Also, since this would take a while to run, it would be great if you could try to run it in any way possible on AWS, so that I can at least use the output to generate a plot or table to the camera ready version (all of the remaining steps should be easy enough for me to run locally)

A primary issue with this code is that since I cannot run docker on our cluster, I cannot test it on a similar environment to AWS.

Checklist before requesting a review

• I have performed a self-review of my code

• Check the correct box. Does this PR contain:

  • Breaking changes
  • New functionality
  • Major changes
  • Minor changes
  • Bug fixes

• Proposed changes are described in the CHANGELOG.md

• CI Tests succeed and look good!