Figure out how best to combine Chandra photometric products

[DavidT3]

Having combined images/exposure maps (if there are multiple observations) is useful, both for visualising and for examining things like the observation coverage of the sky around the source of interest.

Plus XGA does expect there to be combined photometric products for some things.

So we need to make sure that we can add together the photometric products - there is a CIAO task called reproject that will project one ObsID into another's coordinate system, though I think that is more for event lists rather than images (https://cxc.cfa.harvard.edu/ciao/threads/combine/).

I would prefer being able to add images/exposure maps together (like in the emosaic tool for XMM, see the generate.sas.phot section of XGA), but if we have to add reprojected event lists together and use them to make new photometric products then we can. We did do that for eROSITA actually.

Things to think about:

• Researching what tools there are available in CIAO to do this sort of thing.
• Consider that as a backup there is an Astropy-affiliated, well-written, Python package that does this sort of thing (https://github.com/astropy/reproject). Actually one of the outstanding XGA issues (Image products should be able to be added together #1214) is to add a method to the Image class that uses this module to let image instances quickly be added together).
• Especially think about how we can make combined flux-images, those are going to be a real time saver if possible (though I know there isn't yet an XGA class for them) - I think the fluxobs task in Chandra will do it for us (https://cxc.cfa.harvard.edu/ciao/ahelp/flux_obs.html#plist.units) though I haven't read through properly yet.