fix: build HNSW with wrong distance type

[BubbleCal]

No description provided.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 79.67%. Comparing base (0dbfc7c) to head (a2789b1).
Report is 1 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2533   +/-   ##
=======================================
  Coverage   79.66%   79.67%           
=======================================
  Files         207      207           
  Lines       59179    59171    -8     
  Branches    59179    59171    -8     
=======================================
- Hits        47147    47142    -5     
+ Misses       9256     9247    -9     
- Partials     2776     2782    +6     

Flag	Coverage Δ
unittests	79.67% <100.00%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[wjones127]

I'm a little nervous about merging a bug fix without a unit test. Is there a way to catch an issue like this in a test?

[BubbleCal]

I'm a little nervous about merging a bug fix without a unit test. Is there a way to catch an issue like this in a test?

It seems there's no good way to verify it because we won't store the flat storage. I guess we can verify it by recall, and we do have tests for this, but for small dataset it's always enough (>=0.9) even passed wrong distance type.