Apply symops to anisotropy Hamiltonians #99
Description
It would be useful to be able to apply symmetry operations to the axis of anisotropies. This would allow anisotropy in complex crystals such as YIG to be modelled quite easily.
This roughly requires applying the symmetry operations to a vector rather than a point. We discard any translational part and only apply the rotation transformation. It's not clear to me yet if we should do a polar or axial vector operation for the anisotropy axis. I suspect axial because it would be axial for the spins due to the broken time reversal symmetry.
I need to make a decision for whether the symops would be applied by default (a change from the current API) or turned on optionally. Having symops on by default might cause lots of complications in supercells and open systems.