Releasing the option for returning a "precision" matrix for standard_dg_multi calculations