Branch to image using multiple tracks

[awattsup]

Hi processMeerKAT devs,

I thought you might be interested, I have made a script that enables the pipeline to be used to image multiple tracks into one data product.

It works in a similar way to the default pipeline; you build a parameter file with the -BC (--combtracks) flag while supplying a comma-separated list of (M)MS to the -MS flag. These MSs should be calibrated in the case of continuum, or continuum-subtracted in the case of spectral lines.

The pipeline then imports from the existing selfcal scripts to do a blind deconvolution of the sky and make a mask and rms map, and the results are then fed into the existing science_image script.

I tried to keep changes to existing scripts minimal, most changes are to allow the list of MSs to be passed through verification functions successfully.
I have verified it reduces and images data fine, but let me know if there are any unit tests you would like it to run through.

The commits also fix a possible bug in selfcal_part2.py, where the 'loop' variable is referenced but not fed into the function at any point.