expose coupled gradient computation

[anderson2981]

This encapsulates the part that updates the boundaries and computes the gradients for the coupled system. Basically everything that gets done before calling ns_operator and diffusion_operator. This is required when the driver needs the gradients to compute some other mesh properties, e.g. artificial viscosity.

Updates coupled_ns_heat_operator to use the new implementation.

Questions for the review:

• [ ] Is the scope and purpose of the PR clear?
  • [ ] The PR should have a description.
  • [ ] The PR should have a guide if needed (e.g., an ordering).
• [ ] Is every top-level method and class documented? Are things that should be documented actually so?
• [ ] Is the interface understandable? (I.e. can someone figure out what stuff does?) Is it well-defined?
• [ ] Does the implementation do what the docstring claims?
• [ ] Is everything that is implemented covered by tests?
• [ ] Do you see any immediate risks or performance disadvantages with the design? Example: what do interface normals attach to?

[anderson2981]

I wonder if we'd be better off just switching over to the driver-defined coupling as implemented in #902? I think @tulioricci is already using it.

It's certainly possible. I wasn't aware of the other development, so I just tried to rework what was here to support what I wanted to do. At the end of what I did, I don't actually need coupled_ns_operator anymore, I can just directly call the ns_operator and diffusion_operator directly from the driver, but I wanted to preserve other drivers that were using it and have them use the new api.

[majosm]

Is there anything in this PR that we still want to add? Or did #902 mostly cover it?

[anderson2981]

no longer needed.