Correct for nonlinear phase offsets between receivers

[d3v-null]

this thingy
PhaseCorrection.txt

[d3v-null]

basically everything has whitening except RG6_90 and ANY RFOF

[andreww5au]

I would say 'Every cable flavour starting with 'RG6', except for 'RG6_90' and 'everything starting with LMR'. Whenever we define a new cable flavour, it will start with 'RG6', or 'LMR400' (both of which refer to the physical coaxial cable type), or 'RFOF'. For coax cables, the 'RG6' or 'LMR_400' in the cable flavour name is followed by _NNN with the nominal physical length. For fibre cables, we don't bother including the length in the cable flavour name, because it has no effect on the losses as a function of frequency (the fibre has a 'flat' response). If you base your test on just the first few letters of the flavour name, it'll still work if we add 'RFOF-EDA2', or 'LMR400_150', etc, at some point in the future.

[d3v-null]

Instead of the static text file, we will use mwalib 1.5.0 to read the signal chain corrections from the metafits.