Generalize the case of dealing with duplicated column names

We now have multiple modules that have to deal with a csv/other ascii output that has the same column name for multiple columns. Currently we have a workaround implemented in xmatch, but ideally it should be in utils (and refactored to make it work according to the relevant discussions) and to be reused by any modules needing the functionality.

Start by examining the existing duplicate-column-name workaround in xmatch and the relevant discussions referenced by the issue. Identify how the functionality should move into utils and which modules need to reuse it; done means CSV or other ASCII outputs with repeated column names are handled consistently across those modules.

Written by the indexing model from the issue text.