check profile contraction works with magnetic systems.

[pkienzle]

Profile contraction (specifying dA=... when defining the experiment) may be broken for magnetic systems. It needs to be tested and corrected or removed from the program.

dA should allow fine profile steps when building your smoothly varying magnetic profile without incurring too much performance penalty since neighbouring layers will be combined without distorting Q too much below Qmax. At least, that's the idea.

Update the documentation after checking the range of validity and the association between Qmax and dA.