Add colormap option for plot_grid_var.py and grid_var frame generation

[samueltootle]

A small addition to plot_grid_var.py and grid_var to allow changing the colormap of 2D slices via CLI.

[Sbozzolo]

Thanks!

I think we can do a little bit better: we can make all the parameters adjustable.

Here's a couple of ways to do so:

1. Do something like this to set the parameters via command-line. The problem that I see is that by default everything is treated as a string, which might not work.
2. Add a command-line option to specify a config file (possibly controllable via env variable) to set up matplotlib parameters.

The file argparse_heper.py will allow us to make either changes in a relatively clean way for all the files. I think 2 is simpler, but 1 is more flexible because you don't need to create a file to change the parameters. Do you have thoughts about this?

[samueltootle]

Regarding option 1, the string treatment isn't a big deal per say - it's python after all.

Overall option 2 is more appealing. The reason for this is simple: I'm already creating shell files with command lines that i use often or repeatedly. Long commands become increasingly unreadable and not maintainable.

A config file would be a nice alternative so long as it could be done in such a way as to reduce significant refactoring. A compromise would also need to be made in the case that config files allow for more options than the standard CLI interface such that a separate help would need to be available to list the available options without requiring the user to go to the website. That's just my 2¢ at least.

With that said, the current setup with the inclusion of cmap covers 75% of my use cases with the 25% being plots that require custom plots altogether. So, from my side, the level of effort required for either implementation is not really worth the gain from my limited perspective.

If i were to put my effort somewhere it would be Jhuki. Hopefully I have some time after writing my thesis as I have a plan in mind that would automate par file generation for FUKA ID since the python bindings already exist.

[Sbozzolo]

FYI, you can write config files instead of shell scripts (see, --help). You can also intermix setting options from the command line and from config files.

I implemented 2. in next. Now, you can provide a matplotlib configuration file to command-line scripts. In your case, you can create a file named, e.g., ~/.mympl.rc. Then, you can either set up the env variable KBIT_MPL_RC_FILE=~/.mympl.rc (or pass the command-line flag --mpl-rc-file ~/.mympl.rc and add your personal configurations in that file. For example, if you just want to change the colormap, the file will be

image.cmap: viridis

Is this a good solution to your use case?