Better documentation and examples

[mjboos]

The documentation could be improved, especially from the perspective of a newcomer to voxelwise encoding models.
Similarly, it would be great to have concrete examples that show the effect of lagging, the structure of the returned coefficients and preprocessing of fMRI.
It's not strictly necessary to do that with a BIDS dataset, but it would be nicer if we could use one (see #1 )