wobbling sources in ASKAP data

[Elle-WANG]

Adam has a few thoughts:

1. It's more likely to be an ASKAP calibration issue and we probably cannot improve it in imaging step. Might be easier to handle it in post-processing step.
2. facets=2 in principle is unnecessary and a combination usage of facets and wprojplanes=-1 will potential couse w-projection problem (and generate more alias)
3. try tclean usemask = auto-multithresh

A few useful tests

1. try to get residual image and check if the residual image looks reasonable
2. check model image to see if the dots are everywhere or just near the target list (i.e. masking performence)
3. check if those wobbling are direction-dependent or direction-independent (presumbly should be direction-dependent due to ionospiric effect)
4. check bright sources in the field - if they're doing same wobbling behaviour/trend