Apply foreground color to minor ticks

[Carifio24]

There is currently a bug where any minor ticks displayed will not match the user's global settings. (Note that this isn't an issue with the default settings, since the labels are black either way.) One prominent example is the scientific notation tick marks that are sometimes used when displaying an axis in log mode. See below for an example:

This occurs because we only set major tick labels to match the given color in set_foreground_color (see the matplotlib documentation here; the default value for which is "major"). This PR fixes the issue by setting which="both").

[codecov]

Codecov Report

Merging #2305 (eea1ea0) into main (139d88d) will not change coverage.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##             main    #2305   +/-   ##
=======================================
  Coverage   88.06%   88.06%           
=======================================
  Files         247      247           
  Lines       23476    23476           
=======================================
  Hits        20673    20673           
  Misses       2803     2803           

Impacted Files	Coverage Δ
glue/viewers/matplotlib/mpl_axes.py	90.24% <100.00%> (ø)

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

[astrofrog]

Can you move this to 1.6.0? (I am going to work on a way to avoid us having to do this in future)

[Carifio24]

@astrofrog I've rebased and moved the changelog entry to the 1.6.0 section.

[astrofrog]

Could you check the codestyle failure with tox -e codestyle - thanks!

[Carifio24]

Hmm - it looks like the codestyle failure is actually coming from the main branch. Taking a quick look there, the codestyle check went from succeeding to failing after #2300 was merged - although codestyle passed in that PR and didn't change any of the offending files. I think the culprit is the E275 update described here.

[dhomeier]

any of the offending files. I think the culprit is the E275 update described here.

Thanks for identifying this @Carifio24 . The codestyle is fixed in #2308 now.