add support for path_effects to overlay_ax

[Migelo]

When embedding the 2D colormap into the plot itself in full or just a part of it, it can happen that the labels and or ticks of the colormap are not visible due to being very similar in color to the picture itself. I fix this by allowing the user to pass path_effects to add an outline to the labels and ticks.

Here is an example of a problematic plot with $\phi$ not visible on the top and the fixed version below.

You can reproduce the figure above with my fork of the repo with this code snippet.

Happy to modify the functionality in any way :)
Looking forward to hearing your opinion @trygvrad !

import matplotlib.pyplot as plt
import colorstamps
import matplotlib.patheffects as PathEffects

img = colorstamps.helpers.get_random_data()
rgb, stamp = colorstamps.apply_stamp(
    img[:, :, 0],
    img[:, :, 1],
    "funnel",
    vmin_0=-1.2,
    vmax_0=1.2,
    vmin_1=-1,
    vmax_1=1,
)


fig, axes = plt.subplots(2)

# plot the original image
ax = axes[0]
ax.imshow(rgb)

# show colormap as overlay
overlaid_ax = stamp.overlay_ax(
    ax, lower_left_corner=[0.7, 0.85], width=0.2, path_effects=None
)
overlaid_ax.set_ylabel(r"$\phi$")
overlaid_ax.set_xlabel(r"$\omega$")


# plot the fix
ax = axes[1]
ax.imshow(rgb)

# add path effects to make text more readable
path_effects = [PathEffects.withStroke(linewidth=3, foreground="w")]

# show colormap as overlay
overlaid_ax = stamp.overlay_ax(
    ax, lower_left_corner=[0.7, 0.85], width=0.2, path_effects=path_effects
)
overlaid_ax.set_ylabel(r"$\phi$")
overlaid_ax.set_xlabel(r"$\omega$")

[trygvrad]

@Migelo Thank you for the interest in this, I'll take a look and approve it soon.

In other news, I am trying to bring this functionality into matplotlib proper (see issue 14168, and PRs 28454, 28658, [28428]).
In this regard, it would be useful if you could make a comment on the issue (14168) describing your use case (astronomy/microscopy/etc.) and which colormaps from colorstamps you use (and why). That way we can make sure your use case is supported.

[Migelo]

Great!

I am in astro and we sometimes use 2D colormaps to, in a single plot, show two properties of gas, e.g. density and temperature. I used the cut colormap, but I am not particularly attached to it and would be happy to use any better ones that you recommend.

[trygvrad]

I merged this and put a new wheel at pypi (0.1.3)
We should update the docs with a new example (based on the code you provided above).
You can make a new PR if you want, otherwise I will do it when I can find the time :)

[Migelo]

Thanks!
Here is the pull request #3.