Some issues with coordinates docstrings on representation

I found a couple of issues with coordinates docstring on representation using Galactic frame. Codes to reproduce the issues summarized here.

1. That the differentials must be at least 1/time and there is no unitless time (although there is unitless length) should be noted somewhere, and the error message printed when they are not can be improved?
2. Current docstring templating gives wrong information about initializing Galactic frame with the cartesian representation.

[bsipocz]

Well, without any other labelling I think it’s a good assumption is that it’s the effort level of the person doing the labelling :trollface:

[astrojuanlu]

After #9064, (1) was fixed:

In [9]: ICRS( 
   ...:   x=1, y=2, z=3, 
   ...:   v_x=1, v_y=2, v_z=3, 
   ...:   representation_type=r.CartesianRepresentation, 
   ...:   differential_type=r.CartesianDifferential, 
   ...: )                                                                                                                                                                                                                                     
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-9-916f1fc2c13a> in <module>
----> 1 ICRS(
      2   x=1, y=2, z=3,
      3   v_x=1, v_y=2, v_z=3,
      4   representation_type=r.CartesianRepresentation,
      5   differential_type=r.CartesianDifferential,

~/Personal/Astropy/astropy/astropy/coordinates/baseframe.py in __init__(self, copy, representation_type, differential_type, *args, **kwargs)
    580                     current_diff_unit = differential_data._units[f'd_{comp}']
    581                     if not current_diff_unit.is_equivalent(current_repr_unit / u.s):
--> 582                         raise ValueError(
    583                             "Differential data units are not compatible with"
    584                             "time-derivative of representation data units")

ValueError: Differential data units are not compatible withtime-derivative of representation data units

(oops, that missing whitespace!)

and (2) was addressed by adding a note on checking frame_specific_representation_info and representation_info. I think this issue can be closed, if others think there’s still something to be done here please feel free to reopen.