Fix COO symmetrization

[viclafargue]

No description provided.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 80.83%. Comparing base (3c0abf7) to head (71fb5c7).

Additional details and impacted files

@@              Coverage Diff              @@
##           branch-25.04    #2582   +/-   ##
=============================================
  Coverage         80.83%   80.83%           
=============================================
  Files                19       19           
  Lines               527      527           
=============================================
  Hits                426      426           
  Misses              101      101           

☔ 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.

[cjnolet]

Before this is merged, I’d like a comparison done and a small test put in place. This is an important operation on the UMAP side right now and I want to make sure this is actually improving the correctness. Even an MRE in a unit test would be helpful.

[jcrist]

What's the status of this fix? Is this something we'll want to get in for 25.04 (and if so, does anyone have capacity to do the investigation Corey requested above)?

[cjnolet]

Until someone can demonstrate (with an appropriate gtest) that this behavior is causing a problem, I would treat it as a relatively low priority.

[viclafargue]

I will write some tests to check if it actually improves the COO symmetrization.