[daip] CLPLT problems

[Michael Bietenholz]

More dispatches from deep within the bowels of AIPS code:

CLPLT has several problems (31DEC99, about Aug 1, 2000 or so)

1.  It plots +/-0.5 sigma error bars,
2.  which doesn't matter too much, because its calculation of
      the closure sigma is wrong (I think it may mistake
      weight for sigma somewhere)
3.  It also mislabelled the time axis in the case of GST (it
      plots GST in days but labels it as hours).

It also doesn't do a weighted average of the baseline real 
and imaginary's, but I don't know if there is an official
AIPS standard on whether the weights are supposed to be
used everywhere or not.

I have a version, which I am happy to forward to you if you want,
which does a proper weighted average, plots 1-sigma error bars, and
seems to plot error-bars consistent with those plotted by difmap, but
is not that thoroughly tested.

         michael bietenholz