SFF settings in command line & data file

[akelbert]

Currently, to run a secondary field formation (forward modeling only) one has to both add the + SFF line in the data file and use the -E command line option.

Brainstorm some good ways to pass on the SFF request to the solver that would ideally also work for inversion, that would work for MT and other EM configurations, and that would not require editing the data file.

[MiCurry]

If it just needs to be a single flag indicating whether to use spherical or not, could we potentially just add an optional argument similar to what we've done with -v (and the -P I've introduced in #42).

In #42 I updated that optional argument code so it is pretty easy to add additional optional arguments. We could add a '-S' argument perhaps?