[daip] CALIB feature

[Jim Ulvestad]

If you have a VLBA data set that you're trying
to self-calibrate, and the data set is all one
polarization, you get in big trouble by
specifying aparm(3)=1 (average RR and LL) in
CALIB.  In this case, CALIB apparently does
the models, gets to the point where it says
it's writing the SN table, and then goes into
an infinite loop and stays there forever.

Obviously, this is user error for not remembering
how a data set was observed several years ago,
but it would be nice if it croaked a little more
obviously.

jim