Interfacing MiMA with ML codes

[jatkinson1000]

An overarching issue to track progress towatds interfacing various ML models and benchmarking.
Oversees:

• TensorFlow GWP model for MiMA v2.0 #3
• PyTorch GWP model for MiMA v2.0 #4
• Bridging code to configure which GWP is being used #5

where more detail can be found.

@jatkinson1000 - covering #4 and #5

• Adapt cg_drag.f90 to accommodate drag forcing from various different approaches #11
• Create .f90 files with modules & subroutines for directly coupled ML methods #12
• Cannot pass multiple inputs to a model Cambridge-ICCS/FTorch#1

@SimonClifford - covering #3

• Construct a TensorFlow model to match the one in src/shared/pytorch #13
• Create .f90 files with modules & subroutines for directly coupled ML methods #12

We then need to conduct benchmarking:

• Instrument MiMA for ReBench
• Plan a set of runs
• Perform the set of runs