Fix test keybinding for layer actions

[brisvag]

Description

Fixes #5404.

I changed the test so they actually do what they were intended to: ensuring that keybinding existed for each action (rather than having hardcoded numbers like we have now).

To do so I had to programmatically find all relevant actions and shortcuts.

This fixes a few things:

• the result helped me track down a couple of missing shortcuts (which the test was meant to catch) and a quasi-duplicated shortcut for Image.plane.
• I found out what the issue was with problematic side-effects in test functions #5404: layers need to be first instantiated once in order to be correctly populated. Now the test is deterministic. Not sure if we need to somehow fix the behaviour so that instantiation is not required, but that's out of scope for this PR.
• tests are hopefully clearer in intent now that I renamed and restructed them a bit

Type of change

• Bug-fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

References

How has this been tested?

• example: the test suite for my feature covers cases x, y, and z
• example: all tests pass with my change
• example: I check if my changes works with both PySide and PyQt backends
  as there are small differences between the two Qt bindings.

Final checklist:

• My PR is the minimum possible work for the desired functionality
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• I have added tests that prove my fix is effective or that my feature works
• If I included new strings, I have used trans. to make them localizable.
  For more information see our translations guide.

[codecov]

Codecov Report

Merging #5406 (849a00f) into main (b569677) will increase coverage by 0.07%.
The diff coverage is 96.42%.

@@            Coverage Diff             @@
##             main    #5406      +/-   ##
==========================================
+ Coverage   89.02%   89.09%   +0.07%     
==========================================
  Files         595      595              
  Lines       50579    50579              
==========================================
+ Hits        45026    45062      +36     
+ Misses       5553     5517      -36     

Impacted Files	Coverage Δ
napari/utils/shortcuts.py	100.00% <ø> (ø)
napari/_tests/test_viewer.py	98.49% <95.83%> (+2.10%)	⬆️
napari/_qt/widgets/qt_keyboard_settings.py	70.74% <100.00%> (ø)
napari/layers/image/_image_key_bindings.py	78.72% <100.00%> (+9.49%)	⬆️
napari/utils/action_manager.py	81.14% <100.00%> (ø)
napari/layers/shapes/shapes.py	91.39% <0.00%> (+0.21%)	⬆️
napari/layers/image/image.py	95.55% <0.00%> (+0.24%)	⬆️
napari/utils/theme.py	92.68% <0.00%> (+0.60%)	⬆️
napari/utils/info.py	81.44% <0.00%> (+1.03%)	⬆️
... and 4 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

[Czaki]

I like this change. Some plugins will not fail fixed tests.

[Czaki]

Why assignment to _?

[brisvag]

we just need to initialize it, but it's otherwise useless.

[Czaki]

is this xfail required?

[brisvag]

Not sure, I left this untouched from before. I think the this boils down to the same issue below where you need to first instantiate the object for actions to be correctly initialized. I think we can ust remove it?

[brisvag]

Actually, probably better to leave it, it shows that we have some issue with intitialization so we don't forget to solve it at some point. I'd say solving it is out of scope for this PR, though.

[brisvag]

Thanks to the codecov warnings I noticed that it wasn't being run correctly. Now it's fixed! We needed to navigate the full mro() to get any actions.

[andy-sweet]

Sorry, I didn't have time to review this, but I am so happy those magic method counts are gone now!