It is flaky, and our memory "limit" is pretty much arbitrary. Testing
with `tools/memleak.py` shows that memory usage generally rises and only
stabilizes close to 200 figures, while this test only does a warmup of
2 figures. Waiting for equilibrium would take too long for a general
test, so we might move this to a weekly job later.

- Turn each norm into a headed section.
- Link to each norm directly in text.
- Consistently use `Z` as data variable, and pass norms (or `vmin`) in
  the same order for every example.
- Label the colour bars.
- Tweak some descriptions to be clearer.
- Change data for SymLogNorm and use it for the custom norm and
  BoundaryNorm, in order to match their descriptions.

…converters

…es and jpl_units

- deprecated cm api removal documentation
- Corrected formatting issues with documentation
- Removed corresponding type sub in related cm.pyi file and exceptions in mypy-stubtest
- Removed tests, updated documentation with code style for removed functions, moved code to pyplot from cm
- Revert "Removed tests, updated documentation with code style for removed functions, moved code to pyplot from cm"
  This reverts commit a8a4f7f.
- Documetation updates to match code style for deprecated api's
- Removed tests for deprecated api's
- Corrected documentation with full path and verbiage on removals
- Removed handling of override builtin from ColorMaRegistry

EffVer very closely articulates what we are already doing.

Co-authored-by: Tim Hoffmann <2836374+timhoffm@users.noreply.github.com>

This new name was added in Meson 1.1, which is our minimum requirement.

This was added in 0.62, and maybe is a bit more portable.

This doesn't get used for anything effectively, but perhaps meson-python
will start using it eventually.

Closes matplotlib#27774.

Inspired from the discussion in matplotlib#27765: We should visually communicate
that `plot()` covers all three variants: markers only,
line+markers, line-only.
They are visually distinct enough that it's not possible
to infer the variants if you see only one.
In particular, it's important to communicate that you
can draw markers only. We don't want to automatically drive people who
want markers (e.g. some discrete measurements of a dependent variable y
(x)) to scatter because that's the only one showing
 discrete markers in the overview.

Bumps the actions group with 2 updates: [pypa/cibuildwheel](https://github.com/pypa/cibuildwheel) and [codecov/codecov-action](https://github.com/codecov/codecov-action).


Updates `pypa/cibuildwheel` from 2.16.4 to 2.16.5
- [Release notes](https://github.com/pypa/cibuildwheel/releases)
- [Changelog](https://github.com/pypa/cibuildwheel/blob/main/docs/changelog.md)
- [Commits](pypa/cibuildwheel@0b04ab1...ce3fb78)

Updates `codecov/codecov-action` from 3 to 4
- [Release notes](https://github.com/codecov/codecov-action/releases)
- [Changelog](https://github.com/codecov/codecov-action/blob/main/CHANGELOG.md)
- [Commits](codecov/codecov-action@v3...v4)

---
updated-dependencies:
- dependency-name: pypa/cibuildwheel
  dependency-type: direct:production
  update-type: version-update:semver-patch
  dependency-group: actions
- dependency-name: codecov/codecov-action
  dependency-type: direct:production
  update-type: version-update:semver-major
  dependency-group: actions
...

Signed-off-by: dependabot[bot] <support@github.com>

Simplify example: Box plots with custom fill colors

The [original example](https://matplotlib.org/3.8
.0/gallery/statistics/boxplot_color.html) was more complex than need be:

- The two plots only differed in the `notch=True` parameter,
  which is irrelevant for custom fill colors
  -> reduce to one
- rework to a more realistic-like plot

---------

Co-authored-by: Ruth Comer <10599679+rcomer@users.noreply.github.com>

Closes matplotlib#27753

If you are using this and really need unbounded timeouts, explicitly
pass `timeout=None`

On windows on py312 we can not use `.communicate` during process shutdown
because it internally uses Python threads which can no longer be created during
interpreter shutdown.

Closes matplotlib#27437

…ind11.h includes


pybind11 seems decent about including Python.h before any system headers, once you chase three layers of includes in to see that.