[evlatests] doppler tracking

[Barry Clark]

> From: Ken Sowinski <ksowinsk at nrao.edu>
> 
> Might this be related to the fact that we have 1 Hz tuning
> resolution for IFs A and D, but only 2 Hz for IFs B and C?

Yes.  I had been assuming that the A & D Flukes were the masters in
that relationship.  If, instead, B & C are the masters, then the VLA
is inherently incapable of tuning to an odd number of Hz.

Probably best fixed in obs2script.  (Quick and dirty would be to just
drop the Hz digit altogether.)