Fix frame.separation_3d() when frames have velocity data - #6326

Hi there @adrn 👋 - thanks for the pull request! I'm just a friendly 🤖 that checks for issues related to the changelog and making sure that this pull request is milestoned and labelled correctly. This is mainly intended for the maintainers, so if you are not a maintainer you can ignore this, and a maintainer will let you know if any action is required on your part 😃.

Everything looks good from my point of view! 👍

I didn't look in any detail, but do make sure that if differentials are attached, and an obstime is present, this doesn't just work, i.e., the transformation should be done to the other frame before dropping the differentials.

The changes we made happen after the transformation, so I can confirm that if obstimes's are present it doesn't work:

icrs = coord.AltAz(az=15*u.deg, alt=61*u.deg, distance=21*u.pc,
                   pm_az_cosalt=-21*u.mas/u.yr, pm_alt=150*u.mas/u.yr,
                   radial_velocity=100*u.km/u.s, obstime=Time('J2015'))

icrs2 = coord.AltAz(az=15*u.deg, alt=61*u.deg, distance=21*u.pc,
                   pm_az_cosalt=-21*u.mas/u.yr, pm_alt=150*u.mas/u.yr,
                   radial_velocity=100*u.km/u.s, obstime=Time('J2000'))

icrs.separation_3d(icrs2)

raises:

ERROR: NotImplementedError: You cannot transform a frame that has velocities to another frame at a different obstime. If you think this should (or should not) be possible, please comment at https://github.com/astropy/astropy/issues/6280 [astropy.coordinates.baseframe]

@bsipocz I don't think this needs a changelog entry - this fixes a bug based on functionality that only exists in master.

@adrn - Agreed. Basically none of the Affects-dev PRs need a changelog entry by default (except when it's good to have an update on the entry of the original implementation).