Support EQDSK field

[Sichao25]

Support the load of EQDSK file and perform evaluation with uniform grid spline interpolator. Followed the data structure implemented in STOMMS. Close issue #242

[jacobmerson]

@Sichao25 can you take a look at what the configuration issue is? I got a bit behind on approving PRs while asics was offline.

[Sichao25]

@jacobmerson We have two pending updates in self hosted tests #322 and cmake tests #321 .

[jacobmerson]

Thanks I had missed those. They are merged now.

[jacobmerson]

I think we should add SplineFunctionSpace::FromEQDSK that can consume the EQDSKData struct. That would remove a lot of the boiler plate that the user would need to deal with when they want to work with EQDSK files. Internally, that could call the uniform grid constructor.

[Sichao25]

@jacobmerson The new constructor has been implemented.

[jacobmerson]

@usmanriaz07 Are there potentially other fields in the EQDSK that may need to be evaluated later, or is the only field we evaluated in the eqdsk the psi field?

[usmanriaz07]

We do use poloidal current value in the field line tracing. See the following piece of code:
https://github.com/SCOREC/STOMMS/blob/main/stomms_meshgen/src/eqdskData.cc#L197-L203

In ezfit.F90, the relevant function is: https://github.com/SCOREC/STOMMS/blob/main/stomms_meshgen/src/ezfit.F90#L562-L578

Other than poloidal current, we only need psi field.

[jacobmerson]

Is there ever a time where we would have an eqdsk with no data associated with it?

[Sichao25]

There will always be an EQDSKData, but it doesn't have to be created with the helper functions defined here. It can be constructed manually instead of loaded directly from a file. The struct defined here is just for return values.