The alignment tensor objects in the relax data store now support sequential Monte Carlo analyses.
The AlignTensorData.set_sim_num() method was preventing a second Monte Carlo error analysis from
being performed by throwing a RelaxError. The check for previous simulations has been killed.
