* Add back some things [skip ci]

* Remove pytest markers for old async

* Deprecate loaded

* Empty commit to rerun CI

* Empty commit to rerun CI

* Ignore deprecated introspection

force trigger CI

…n and shortcut (napari#5636)

# Description

The current trash-icon delete button in the viewer when clicked deletes
selected layers, but is also supposed to accept drag-n-drop events,
making it work like the Trash on say macOS (and other OS with a similar
desktop model).
However, this behavior was broken in
napari#2441

This (broken) behavior makes no sense, because napari *doesn't use* a
desktop model, so the Trash icon is not a folder, like the Trash bin in
a OS, a temporary storage for unwanted items. You can't open the napari
trash button to restore the layers you drag there.

The Trash icon in napari should be simply a button—and is stylized as
one—that responds to a single-click to delete selected layers.

This PR removes the broken behavior—eliminating the source of error. In
this way it simplifies the code, to treat the delete button the same as
other viewer buttons (QtViewerPushButton). as a result, the delete
button now also has a registered action and keybinding that is editable
in the Preferences, rather than hard-coded.


## Type of change
<!-- Please delete options that are not relevant. -->
- [x] Bug-fix (non-breaking change which fixes an issue)
- [ ] New feature (non-breaking change which adds functionality)
- [x] Breaking change (fix or feature that would cause existing
functionality to not work as expected): it's been broken for 2 years
without being reported...
- [x] This change requires a documentation update: docs PR:
napari/docs#123

# References
closes napari#5629
See also discussion on zulip:
https://napari.zulipchat.com/#narrow/stream/212875-general/topic/Drag.20layer.20to.20trash

# How has this been tested?
<!-- Please describe the tests that you ran to verify your changes. -->
- [ ] 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:
- [x] My PR is the minimum possible work for the desired functionality
- [ ] I have commented my code, particularly in hard-to-understand areas
- [x] I have made corresponding changes to the documentation: docs PR
napari/docs#123
- [ ] 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](https://napari.org/developers/translations.html).

…d labels width (napari#5640)

# Description

Old behavior:

![GIF showing the old behavior of the viewer slider widgets axis label
resize. The labels take the width of the longest axis label element even
if it is not being displayed. The space for the axis labels is the same
even if the current axis label text is shorter than the space
available](https://user-images.githubusercontent.com/16781833/225688465-df3804ce-2b37-423a-9ba4-1343db83f03c.gif)

New behavior:

![GIF showing the new behavior introduced with the PR for the viewer
slider widgets axis label resize. The labels take the width of the
longest axis label element that is being displayed. The alignment
between multiple slider widgets is preserved but changing the axis being
displayed can make the axis labels to take more or less
space](https://user-images.githubusercontent.com/16781833/225688739-96db3175-40dd-49db-8745-6156bb2a13d5.gif)

## Type of change
- [x] Bug-fix (non-breaking change which fixes an issue)

# References
closes napari#5521 

# How has this been tested?
- [x] all tests pass with my change
- [x] I check if my changes works with both PySide and PyQt backends
      as there are small differences between the two Qt bindings.  

## Final checklist:
- [x] My PR is the minimum possible work for the desired functionality
- [x] I have commented my code, particularly in hard-to-understand areas

---------

Co-authored-by: Andy Sweet <andrew.d.sweet@gmail.com>

This will be an error in Python 3.11, See CI in napari#5439.

Send this a separate commit to make review easier (here and in napari#5439
that should in the end just enable CI on 3.11)

No tests added, as it will become an error in future Python version
anyway.

# Description
This introduces a `feature_defaults` init parameter and property setter
for the three layer types that support that (points, shapes, and
vectors). It also fixes some related issues that occur when adding
feature-derived colors when modifying the feature defaults in-place.

Note that the `feature_defaults` is deliberately a little different to
the `current_properties` setter, as the latter also has the side-effect
of setting the feature values of the selected elements to the new
current property values. Also, using the `feature_defaults` setter with
keys/columns that are extra/missing compared to `features` will result
in an error.

## Type of change
- [x] Bug-fix (non-breaking change which fixes an issue)
- [x] New feature (non-breaking change which adds functionality)

# References
Closes napari#5632
Closes napari#5634

# How has this been tested?
- [x] added new tests to cover bugs and setter
- [x] all tests pass with my change

This skips version 6.4.3 of pyside6 when testing to avoid the problems
described in napari#5657. It does not close that issue, but at least reduces
noise on the PR tests.

- [x] Bug-fix (non-breaking change which fixes an issue)

Related to napari#5657

- [x] all tests pass with my change

Co-authored-by: Grzegorz Bokota <bokota+github@gmail.com>

# Description

This PR stores the layer visibility states prior to toggling layers and
then checks against it to ensure that layers are toggled relative to
that initial state. This makes sure they are not double toggled if
linked layers are involved.
Also added a test for the interaction between linked layers and
toggle_visibility that fails without this PR but now passes.

## Type of change
<!-- Please delete options that are not relevant. -->
- [x] Bug-fix (non-breaking change which fixes an issue)

# References
Closes napari#5655 

# How has this been tested?
<!-- Please describe the tests that you ran to verify your changes. -->
- [x] I added a test that fails on main but passes here
- [x] all tests pass with my change 

## Final checklist:
- [x] My PR is the minimum possible work for the desired functionality

Co-authored-by: Grzegorz Bokota <bokota+github@gmail.com>

…ues (napari#5635)

# Description
This fixes an off-by-one error on Surface layers extent data, which was
causing an IndexError slicing that dimension.

## Type of change
- [x] Bug-fix (non-breaking change which fixes an issue)

# References
closes napari#5620
Disussed in [community meeting on
2023-03-15](https://hackmd.io/BXWDZ3i8Q6OAEASrkaSNIQ?view#2023-03-15)

# How has this been tested?
- [x] all tests pass with my change (though I had to change two of them)
- [x] thanks to @brisvag who [tested to make sure the mesh was in the
correct spot wrt an
image](napari#5620 (comment))

## Final checklist:
- [x] My PR is the minimum possible work for the desired functionality
- [x] I have commented my code, particularly in hard-to-understand areas
- [x] I have made corresponding changes to the documentation
- [x] I have added tests that prove my fix is effective or that my
feature works
- [x] If I included new strings, I have used `trans.` to make them
localizable.
For more information see our [translations
guide](https://napari.org/developers/translations.html).

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

<!-- What does this pull request (PR) do? Why is it necessary? -->
<!-- Tell us about your new feature, improvement, or fix! -->
<!-- If your change includes user interface changes, please add an
image, or an animation "An image is worth a thousand words!" -->
<!-- You can use https://www.cockos.com/licecap/ or similar to create
animations -->

* `INP` flake8-no-pep420 - The pep 420 is something that part of us may
remember as the reason why napari stop working at some moment because of
some problematic files in `site-packages`.

 * `PYI` flake8-pyi - better type annotation
* `Q`, flake8-quotes - enforce quoting, but the rule Q000 is disabled
(see napari#5589).
* `RSE` # flake8-raise - simplify raise code by remove `()` when
obsolete.
* `RET` # flake8-return - simplify code readability by forcing proper
return schema.
 * `TID` # flake8-tidy-imports - replacement for absolutify imports
* `TRY` # tryceratops - enforce rules on exception raises like keeping
in try only this that could raise exception (see why it could create
problems in napari#5577).
* `ICN` # flake8-import-conventions - enforce import convention for
alias (like `import numpy as np`)
 * `RUF`# ruff specific rules -

<!-- Please delete options that are not relevant. -->
- [x] Maintenance

<!-- What resources, documentation, and guides were used in the creation
of this PR? -->
<!-- If this is a bug-fix or otherwise resolves an issue, reference it
here with "closes #(issue)" -->

<!-- Please describe the tests that you ran to verify your changes. -->
- [ ] 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.

- [ ] 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](https://napari.org/developers/translations.html).

The script was not handling string with more than one letter prefix, in
more details, it was trimming only the first letter of the prefix, and
thus when evaluating for example a rf"" f-string, the f would not be
removed, and the eval would look try to interpolate the ftrings in the
context of the function that then would fail.

Now we keep the prefix, but remove "f" from it.

As a second side effect, this also mean that we now distinguish
bytestrings (`b""`), anc can just skip over those as they should never
be translated.

Another addition in the interactive translation updating script, is that
it now print at the beginning an estimate of the number of places that
are missing translation or need to be added to the ignore file.

The translation tests will still fail, but this is because there are
missing translation/entries in the ignore file.

I did not update he ignore file yet to keep things reviewable.

# Description
<!-- What does this pull request (PR) do? Why is it necessary? -->
<!-- Tell us about your new feature, improvement, or fix! -->
<!-- If your change includes user interface changes, please add an
image, or an animation "An image is worth a thousand words!" -->
<!-- You can use https://www.cockos.com/licecap/ or similar to create
animations -->

## Type of change
<!-- Please delete options that are not relevant. -->
- [ ] 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
<!-- What resources, documentation, and guides were used in the creation
of this PR? -->
<!-- If this is a bug-fix or otherwise resolves an issue, reference it
here with "closes #(issue)" -->

# How has this been tested?
<!-- Please describe the tests that you ran to verify your changes. -->
- [ ] 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](https://napari.org/developers/translations.html).

In many cases we actually want to only catch ModuleNotFoundError to
indicate that the module is not installed, catching/suppressing
ImportError also tend to hide deeper import issues and may make things
harder to debug.

…apari#5669)

# Description

This fixes issues with Pan/Zoom buttons (layer controls) not working and
the spacebar keybinding to activate pan/zoom mode also not working.
In napari#4894 there was some refactoring
which introduced the bugs, see my investigation here:
napari#5654

This PR fixes the typo, lack of proper layer types, and now moves the
`hold_for_pan_zoom` to be a Viewer keybinding. As a result, this
keybinding isn't hard-coded anymore, but is now settable in the
Preferences > Shortcuts.

## Type of change
<!-- Please delete options that are not relevant. -->
- [x] Bug-fix (non-breaking change which fixes an issue)
- [x] 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
closes napari#5654

# How has this been tested?
Good question! All tests were passing before this PR, despite the fact
that things were broken. So I'm open to suggestions how to improve tests
so we catch stuff better.


## Final checklist:
- [x] 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](https://napari.org/developers/translations.html).

# Description
closes napari#5677

I tracked down the source of the added Transform and Pan/zoom in the
Viewer keybindings.
Those are actions/keybinds added in
napari#4894 for the Tracks layer. But
Tracks layer wasn't added as an option in the shortcuts table, so they
appear in the default (Viewer).
With this PR, I import Tracks and add it to the dropdown so Tracks has
it's own table and now the actions appear in the proper table, rather
than under Viewer keybindings.

<img width="1198" alt="image"
src="https://user-images.githubusercontent.com/76622105/228287187-189cb745-fb6e-4c31-93f1-6ba22e965a56.png">

<img width="1198" alt="image"
src="https://user-images.githubusercontent.com/76622105/228287227-da2642b7-8f76-4451-bdfa-e6801f47400a.png">



## Type of change
<!-- Please delete options that are not relevant. -->
- [x] 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?
<!-- Please describe the tests that you ran to verify your changes. -->
- [ ] example: the test suite for my feature covers cases x, y, and z
- [x] 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:
- [x] 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](https://napari.org/developers/translations.html).

…apari#5682)

# Description
Alternative to napari#5662. This is actually addressing the issue, instead of
inventing new problems to solve :P

This fixes the original problem; this will now work as intended and
correctly update the dropdown menu and the global `AVAILABLE_COLORMAPS`
dict.


```
import napari
import numpy as np
v = napari.Viewer()
i = v.add_image(np.random.rand(10, 10, 10), colormap=np.random.rand(5, 3))
s = v.add_surface((np.random.rand(5, 3) * 50, np.random.choice(range(5), (3, 3))), colormap=np.random.rand(5, 3))
```


A side effect of the current version is that instead of `custom`, new
unnamed colormaps are named `[unnamed colormap i]` where `i` is an
incremental index. We can change the default to `custom`, if preferred,
but for now I opted for minimal code changes for easier review.

Note that a notable change (read: bug fix) of this PR is that the code
above generate 2 different colormaps, instead of overwriting a single
`custom` colormap.

cc @andy-sweet 
<!-- What does this pull request (PR) do? Why is it necessary? -->
<!-- Tell us about your new feature, improvement, or fix! -->
<!-- If your change includes user interface changes, please add an
image, or an animation "An image is worth a thousand words!" -->
<!-- You can use https://www.cockos.com/licecap/ or similar to create
animations -->

## Type of change
<!-- Please delete options that are not relevant. -->
- [x] 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
<!-- What resources, documentation, and guides were used in the creation
of this PR? -->
<!-- If this is a bug-fix or otherwise resolves an issue, reference it
here with "closes #(issue)" -->

# How has this been tested?
<!-- Please describe the tests that you ran to verify your changes. -->
- [ ] 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](https://napari.org/developers/translations.html).

# Description

Currently, when PRs have long and detailed descriptions (a good
thing!!!), it's
hard to find the reference to the issue(s) that they close. This updates
the
template so that those issues are linked right at the top, before the
description.

I think a lot of the stuff at the bottom of the PR template is cruft (it
was
created a long time ago), but we can discuss that in separate
issues/PRs.

Now that `lazy_loader` is being released as a package, it may be easier
to depend on it than to vendor the code.

@tlambert03 has implemented a feature in `lazy_loader` that allows lazy
loading, while also letting type checkers do their jobs. See
https://scientific-python.org/specs/spec-0001/#type-checkers

I could transition the napari lazy imports to the stub loader in a next
PR, if that's desired.

…ng) (napari#5672)

# Description

This adds an action analogous to the current `new_label` action (current
keybind `M`) to set the current label to background (0). The default
keybind has been set to `B` because it's close to space and close M,
plus B for `background`. Open to other suggestions.
This is sort of similar to `activate_labels_erase_mode` but in a lot of
ways more powerful, because you can use the `fill` tool to erase
contiguous areas.

## Type of change
<!-- Please delete options that are not relevant. -->
- [ ] Bug-fix (non-breaking change which fixes an issue)
- [x] 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
Closes napari#5666
First came up in a docs PR:
napari/docs#125 (comment)

# How has this been tested?
<!-- Please describe the tests that you ran to verify your changes. -->
- [x] example: I added a test
- [x] 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:
- [x] 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: will
update napari/docs#125 once the default keybind
is settled.
- [ ] 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](https://napari.org/developers/translations.html).

…verything back in

* Add back some things [skip ci]

* Remove pytest markers for old async

* Deprecate loaded

* Empty commit to rerun CI

* Empty commit to rerun CI

* Ignore deprecated introspection