* Removed templates from various functions, as we can only produce double models anyway. * Added methods to compute state rewards and state labelings * Added toModel() routine