[daip] Adverb DOFIT

[Eric Greisen]

When some user made a query about this adverb in CALIB I examined its
use there.  I discovered that it was attempting to say (if antenna i
was not to be fit) that the gain of i is (1,0) and data from all
baselines to i may be used in fitting the gains of antennas j.  I
suppose the goal was to use more data but not be dragged by
questionable antennas.  But forcing a questionable antenna to (1,0)
which is known to be wrong is worse than letting it be fit.  The
ANTENNAS adverb allows antenna i to be omitted altogether including
all baselines i-j.  I changed CALIB to fit all gains but store (1,0)
for DOFIT false antennas but I think the option should be removed
entirely.

I now notice that the option also appears in FRING and KRING.  If the
option is implemented similarly, then it too must be in error I
suspect.  Does anyone know how it is implemented there?

Do you know of any reason to have this adverb anywhere?

Eric