Add flux column keyword to Spectrum.read

[rosteen]

Follow-up from #1143 : it would be nice in Spectrum1D.read() to have a keyword specifying that you want the flux column rather than surface brightness even if the SRCTYPE is "EXTENDED" for JWST (and probably other) data, and vice-versa.

[keflavich]

I'd like to bump this. There are many spectra in the JWST archive (everything I'm interested in using) that have the "EXTENDED" tag even though the observed object is a point source or my intent is to treat it as a point source (i.e., integrate over the aperture).

[keflavich]

See also https://github.com/spacetelescope/jwst-pipeline-notebooks/issues/69, which highlights the need for column selection as a requirement for JWST pipelines.

[kelle]

maybe we need a optional flux_column_name argument.

[rosteen]

I have an internal ticket to add this, I'll bump it up in the priority queue.