Enables setting hatch linewidth in Patches and Collections, also fixes setting hatch linewidth by rcParams

[Impaler343]

PR summary

Fixes #21108 using the patch provided by @anntzer which I modified to pass tests. Need to write tests for:

• [x] Setting hatch linewidth in a Patch
• [x] Setting hatch linewidth in a Collection

PR checklist

• [x] "closes #0000" is in the body of the PR description to link the related issue
• [x] new and changed code is tested
• [N/A] Plotting related features are demonstrated in an example
• [N/A] New Features and API Changes are noted with a directive and release note
• [N/A] Documentation complies with general and docstring guidelines

[Impaler343]

I don't know why docs and CI are failing, could somebody please check this out?

[rcomer]

Hi @Impaler343, I'm not certain but I think the docs failure may be related to something that has been fixed on main. Could you try rebasing?

git fetch upstream
git rebase upstream/main

Appveyor is showing the same failure on all PRs right now, so I wouldn't worry about that one.

[Impaler343]

Do I also need to squash?

[tacaswell]

There is enough exploration/experimentation in here it is probably worth squashing to a handful of commits.

[tacaswell]

I think this is like 80% of the way there and looking pretty good!

[Impaler343]

Can we get a feature for check_figures_equal() similar to image_comparison()'s remove_text feature? As the former is a more economical way to test, having it for check_figures_equal() would be a good idea. Or is it already available, and I don't see it?

[oscargus]

Can we get a feature for check_figures_equal() similar to image_comparison()'s remove_text feature?

I do not think this is required? The reason remove_text exists is that the font files may be different on different machine. However, check_figures_equal() are executed on the same machine, so should never be a problem. Am I missing something?

[Impaler343]

Can we get a feature for check_figures_equal() similar to image_comparison()'s remove_text feature?

I do not think this is required? The reason remove_text exists is that the font files may be different on different machine. However, check_figures_equal() are executed on the same machine, so should never be a problem. Am I missing something?

I suggested this as it is a quick plugin to make tests run faster by focussing on the important part of the plot, by removing axes, labels and ticks, and this most probably should make a significant difference while running a large amount of tests

[Impaler343]

Have removed the if conditions as they are unreachable:

• hatchstyle cannot be None as a non-None value of hatch is passed in the function hatch_cmd before being called.

On another note, a lot of lines in backend_bases is uncovered, and I don't think I can write tests for all of them. What to do?

[Impaler343]

Anything else to add or change?

[QuLogic]

I think the doc build failure is unrelated; you may have to rebase/merge to fix it.