Inconsistency of the second moments of MCS centroid measurement

During the MCS alignment, we used MPFIT for fitting 2D gaussian function to determine the spot sizes. We were expecting to use the second moments from database table `mcs_data`. However, the FWHMs calculated from second moments are not consistent with MPFIT of Python and IDL version. We should double check the consistency, so that we can use those information for MCS image quality check.