Accept more formats for the `groups` parameter

[richford]

The common groups parameter currently accepts the explicit standard group format (i.e. a list of numpy arrays where each array represents a group and each element of the array represents a feature index that belongs to that group) or None, in which case all features are assigned to one group.

We should also allow users to input a single integer s, in which case we will create contiguous groups of size s until all features are accounted for.

@arokem, @mnarayan Any other input behaviors we'd like to try?

[richford]

From @jyeatman, we should also allow a vector of integers, where the length of the vector matches the number of columns and the value of each element is a group ID.

[JohannesWiesner]

I would highly appreciate this feature! At least for me, producing these nested arrays of indices is quite tedious. In the cca-zoo package you also simply input a list-like object of integers:

https://cca-zoo.readthedocs.io/en/latest/api/models.html#cca_zoo.models.GRCCA

And at least in the field of neuroimaging, I guess the standard-scenario is to have some sort of table that maps brain region indices to groups of higher order regions (which again map to higher order regions and so forth).

E.g.:

Would be much more convenient to do something like:

estimator = groupyr.SGL(...,groups=atlas_info['region_idx'])