sh2peaks direction sets #3242
Description
It seems that sh2peaks currently defines its own default set of directions, despite there being a standardised pre-defined set of an identical size.
It also occurred to me that the default number of directions to use should ideally grow as a function of lmax of the input data. Eg. for FODs with lmax=16 one might need more than 60 seed directions to find an aPSF / delta function.