Weighted position uncertainties are the same in both RA and Dec (and are possibly too small?)

[ddobie]

The position uncertainties are different in the measurements file, so it's not a fundamental source finder problem. This is present on the website and in the parquet files, so the problem is somewhere in the calculation.

Skimming over the code, there's no obvious typos that assign things to the wrong values... but that's the only explanation I can think of. Need to investigate.

Thanks to @IrisdeRuiter for helping point this out