Adapt a coastal or inland water bodies model from dglc

[NickSzapiro-NOAA]

trafficstars

There are fjords/sounds/lakes/... that are not represented well at given resolutions (see https://github.com/NOAA-EMC/global-workflow/issues/3343 for one example)

Increasing resolution is not always practical. It's also not necessarily easy for land component to tile these, particularly if domain should complement ocean.

It seems like dglc already has separate domains and vertical dimension and sits well along with the other components. Maybe it can be adapted for less-frozen waters as well?

This is intended more as a discussion than issue...maybe @mvertens or others have thoughts?