Add create surface from labels

[arnaudbore]

Quick description

First attempt to bring some of the nice scripts from Etienne to scilpy.
I choose to reduce the scope of the original script to fit scilpy requirements
I kept the expected behavior of the script used in nf-set.

...

Type of change

Check the relevant options.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Provide data, screenshots, command line to test (if relevant)

Use scilpy atlas test data
...

Checklist

• My code follows the style guidelines of this project (run autopep8)
• I added relevant citations to scripts, modules and functions docstrings and descriptions
• I have performed a self-review of my code
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I moved all functions from the script file (except the argparser and main) to scilpy modules
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes

[pep8speaks]

Hello @arnaudbore, Thank you for updating !

• In the file scripts/scil_surface_from_labels.py:

Line 137:80: E501 line too long (86 > 79 characters)

[frheault]

I believe this script should work with a single mask (uint8) as well as labels (uint16), Like if I want a glass brain right now I have to artificially convert my BET mask to unint16

[frheault]

I say the default should open in MI-Brain correctly on top of the NIFTI it was generated for. Should switch to --vtk2vox

[frheault]

As of now, this is crashing unless given (so I would use required at least).
But since this is a script <from_labels> I think the default should be:
generate all the labels (from np.unique() then the script does the for loop itself) into an output folder, if the user wants just a few then --indices uses just one.
similar to the split labels script we could use an --out_dir OR --out_suffix .

This is my recommendation because if you run this on Freesurfer atlas, most people don't know the labels one by one and they probably want to generate a bunch of independent surface. And if they want to merge a few labels into a single mask, I propose that @arnaudbore you reuse the same logic as in scil_tractogram_filter_with_roi.py (like --indices 1 2 3 4 (do only 4 surfaces instead of all) VS --merge_indices 1:6 9 10:15 (do a single surface from a bunch of labels) and these two would be mutually exclusive options (default being all labels as multiple independent surfaces)

[frheault]

index?

--opening --smooth