enhance plot_shapes

[lopollar]

plot_shapes works quite fast, but some extra functionality would be handy:

• Add the option to plot without image
• add the option to plot on previous figure (give ax as an input of the function)
• add the option that if multiple columns are given multiple plots are made
• add the option that if different crds are given, different plots are made

[ArneDefauw]

the napari_sparrow.pl.plot_shapes function now uses the napari_sparrow.pl._plot._plot_shapes function under the hood. The _plot_shapes function returns a matplotlib.pyplot.Axes object that can be used to easily make subplots.

napari_sparrow.pl.plot_shapes function keeps same functionality as before, but now it is possible to specify multiple image layers ,shapes_layers or channels as lists. Combination of image_layer and shapes_layers are plotted as columns of the figure, channels as rows.

In this way, both napari_sparrow.pl.plot_segment and napari_sparrow.pl.transcript_density can both use the same underlying napari_sparrow.pl.plot_shapes function, which simplifies the code.

Do we still implement the other extra features? With the availability of napari_sparrow.pl._plot._plot_shapes, the following two should be straightforward to implement:

-add the option that if multiple columns are given multiple plots are made
-add the option that if different crds are given, different plots are made

Do we want to generate one plot with subplots, or multiple plots?

[lopollar]

I think one plot with subplots is the default and makes it easier to compare.
When people want them separately, they can always just ask to run it 5 times in a row.