fix: propagate mmcore instance to various widget constructors

[tlambert03]

This follows up on #463, adding a test that catches a number of additional missed cases where the core wasn't passed through

It adds a very useful fixture called assert_max_instance_depth ... which I used to catch most of these errors. But there are still a couple errors remaining in test_useq_core_widgets.py and test_config_groups_widgets.py ... so it's commented out for now.

cc @gselzer if you're curious

[codecov]

Codecov Report

❌ Patch coverage is 70.00000% with 9 lines in your changes missing coverage. Please review.
✅ Project coverage is 87.80%. Comparing base (eb950ad) to head (f2d1bf0).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
..._group_preset_widget/_group_preset_table_widget.py	0.00%	4 Missing ⚠️
src/pymmcore_widgets/control/_rois/_vispy.py	0.00%	3 Missing ⚠️
..._presets/_objectives_pixel_configuration_widget.py	50.00%	1 Missing ⚠️
src/pymmcore_widgets/control/_objective_widget.py	0.00%	1 Missing ⚠️

❌ Your patch status has failed because the patch coverage (70.00%) is below the target coverage (85.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #469      +/-   ##
==========================================
+ Coverage   87.77%   87.80%   +0.02%     
==========================================
  Files         106      106              
  Lines       11668    11665       -3     
==========================================
  Hits        10242    10242              
+ Misses       1426     1423       -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.

[gselzer]

Looks nice...it'd be great if we can fix the remaining test failures such that the fixture could be uncommented. I filed a PR to that branch doing so for one of the failing test files, and could look at the other if that'd be helpful.

[tlambert03]

thanks @gselzer. Yes it would be nice, but these are tricky cases for a reason, and I'm not yet ready to go down the path that your PR goes down. The whole point of the new config model was to not need a constant core connection, but rather convert the core into pure model objects (so, we shouldn't need a constant core connection there just to get an icon... it's a leaky abstraction at the moment, but the fix is to grab the device type at model creation time, not to preserve the core). we can get to it soon enough. This PR is already a significant improvement, and I don't want to hold up the "good" for the "perfect".