Return factors and loadings as matrices instead of data frames

[willtownes]

As pointed out by willtownes/scrna2019#5 glmpca is returning factors as a data frame instead of a matrix (as the documentation claims), which causes downstream problems converting SingleCellExperiment objects to Seurat.

1. Expose an option to allow the user to decide whether to return matrix or data frame
2. Change the default to return matrices instead of data frames

The offending place in the code is here:
https://github.com/willtownes/glmpca/blob/master/R/postprocess.R#L86

[willtownes]

Tagging @liuzhe93 so she can keep track of progress here.

[liuzhe93]

Tagging @liuzhe93 so she can keep track of progress here.

ok, thanks a lot. Thanks for your kind help.