Feature request: compiling multiple model_mods into dart

Use case

What are you trying to accomplish? Sometimes there may already be a way to do what you need.

Use multiple model_mods with dart programs to have more than one model in the state

Is your feature request related to a problem?

PSIF project on observations and forward operators
Forward operators which use state from multiple models, e.g. ocean + sea ice.

Describe your preferred solution

Please provide enough detail here to help people have a discussion about your proposal

Here is xmodel, which builds with ./quickbuild.sh

model_config.sh - where the models to be included are declared

./show_config.sh prints out the xmodel configuration.

https://github.com/hkershaw-brown/DART/tree/xmodel/models/xmodel

README.md
markdown docs at the moment, not sure whether these instructions are going in Jupyter notebooks.

Depends on "Route model_mod calls for DART programs though assim_model_mod" https://github.com/NCAR/DART/pull/1069

Describe any alternatives you have considered

List any alternative solutions or workarounds you have considered.
Not using DART for forward operators, using some external tool.

note - I added a note about this as a discussion in https://github.com/NCAR/DART/discussions/1084 but fleshed out feature request here.

Start with models/xmodel/README.md and the example scripts: model_config.sh declares included models, while quickbuild.sh builds and show_config.sh prints the configuration. Read the dependency on PR #1069 before assessing the compilation path. Done means a DART program can use multiple model_mods and the setup instructions describe the workflow.