Fixed depricated changing of cmap

[vetlewi]

In Matrix.plot() the global cmap is changed to set zero to white. Modifications to global cmap is being deprecated by matplotlib, thus a copy should be made before any changes can be made.

I've added an explicit keyword argument for cmap so that if a user given cmap is given it also will get zero counts set to white. Let me know what you think about it. It might make sense to not do that as well, or have a flag in addition to dictate if zero should be white or not.

A little more context:

WARNING:MainThread:py.warnings:ompy/ompy/matrix.py:329: MatplotlibDeprecationWarning: 
You are modifying the state of a globally registered colormap.
In future versions, you will not be able to modify a registered colormap in-place.
To remove this warning, you can make a copy of the colormap first.
cmap = copy.copy(mpl.cm.get_cmap("viridis"))

[fzeiser]

Couldn't one add a cmap via the kwargs already before?

[vetlewi]

To avoid adding extra arguments one could also just set the default value for cmap in kwargs:

...
current_cmap = copy.copy(cm.get_cmap())
current_cmap.set_bad(color='white')
kwargs.setdefault('cmap', current_cmap)
...

[fzeiser]

To avoid adding extra arguments one could also just set the default value for cmap in kwargs:

...
current_cmap = copy.copy(cm.get_cmap())
current_cmap.set_bad(color='white')
kwargs.setdefault('cmap', current_cmap)
...

Not a big difference, but I like this better. It easily starts to get messy if one provides lots of default kwargs

[vetlewi]

Think ready for merge 👍

[fzeiser]

Yes, just trying to see why Travis is not asked for the check by default

[fzeiser]

Travis is triggered now :+1: I've had issues which I through I had fixed this week but migrated to travis-ci.com now. Maybe this fixed it? Anyhow, thanks for the PR.