[PROJ] Setting the GEOIDMODEL in the CRS short id

[Kristian Evers]

> I agree with Greg. Transformation is part of the metadata and not part of the CRS.
> It is a good idea to include information on the transformation history in the metadata, not only for height but also for horizontal coordinates. It would help if software would do that automatically in a standardised way.

This explains why the Danish height system DVR90 is now in EPSG as a datum ensemble consisting of 

1. DVR90(2001), heights from levelling
2. DVR90(2002), based on 2002-geoid
3. DVR90(2013), based on 2013-geoid
4. DVR90(2023), based on 2023-geoid

By using the specific member of the ensemble you *can* communicate how those heights came to be. If you don't know (or don't care), you can just use the ensemble.

This is not a perfect solution but I think it is a good as it gets within the boundaries of current standards. And done this way the software can do this both automatically and standardised.

/Kristian