From #7165 (comment):

Why is there a for-loop at all?

I see the for loop was added in #4347 here https://github.com/napari/napari/pull/4347/files#diff-1807f13d5d575e770c4b24ce95223949ab74d5492ec5eb1a4f71adfcf2e2531c @DragaDoncila any recollection as to why?

I've removed the loop but happy to add it back...

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 92.92%. Comparing base (9a9e7e8) to head (6859c71).

@@            Coverage Diff             @@
##             main    #7172      +/-   ##
==========================================
- Coverage   93.01%   92.92%   -0.10%     
==========================================
  Files         619      620       +1     
  Lines       56757    56762       +5     
==========================================
- Hits        52792    52744      -48     
- Misses       3965     4018      +53     

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

#7172 (comment)

@DragaDoncila says it would be safer to keep the loop but we can add it back if problems 😅😅

I'm just concerned that it's been there for a while now and I won't have time to properly look at this PR and the one that introduced it until I'm back, which won't be until Friday next week at the absolute earliest. If the PR without changing the loop is confirmed to fix the reported issue, I'd prefer to keep this PR to very minimal changes.

I think this PR is the minimal change.

Anyway if anything else breaks with 0.5.2 you'll be back just in time to fix for 0.5.3. 😂 (Less flippantly: with our faster release cycle bugs are less critical, this one closes an existing bug and adds a regression test for it, so we can fix any new bugs more confidently in the future. And I don't think this will introduce new bugs, I think it's the right way to think about this — find a single path through the code, rather than add even more branches at different levels.)