spatially-conditioned permutations

[knaaptime]

resolves #244

following the discussion linked above, this changes the permutation logic for the random_label test in the comparative inference case. Specifically, rather than permute group membership (of all rows, including geometry), it permutes the group values, leaving geometry in place. This allows reasonable computation of all spatial indices

[codecov]

Codecov Report

❌ Patch coverage is 60.00000% with 4 lines in your changes missing coverage. Please review.
✅ Project coverage is 82.48%. Comparing base (9a8cbd3) to head (19105e5).
⚠️ Report is 3 commits behind head on main.

Files with missing lines	Patch %	Lines
segregation/inference/comparative.py	60.00%	4 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #261      +/-   ##
==========================================
- Coverage   82.58%   82.48%   -0.10%     
==========================================
  Files          66       66              
  Lines        2302     2306       +4     
==========================================
+ Hits         1901     1902       +1     
- Misses        401      404       +3     

☔ 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.

[renanxcortes]

If this PR gets merged, then #245 gets deprecated (as it would be contemplated in this PR) and can be closed, right?

[knaaptime]

yeah, this would supercede #245