naming: measure and modality

[nsheff]

Right now we're using the terms 'measure' and 'modality' to refer to the same thing -- should we standardize that?

[dampierch]

Current consensus as I understand it:

Modality will refer to the type of omics data (e.g. RNA-seq, ATAC-seq) in question. It will be an object of type character string. I don't understand dictionaries enough to know where it fits in the dictionary structure.

Measure will refer to the AnnData object that stores variable and observation annotations along with their associated matrix of values for a given modality. It will therefore be dropped from our module, because it is equivalent to the AnnData object itself. Internal to our new object of class MultiAnnData, we will refer to the multiple AnnData objects as a unit (i.e. to the MultiAnnData object itself) with the variable adatas.