Add rhs argument to enable "poisson interpolation" in laplace_interpo…

[Huite]

This would allow us to do fancy downscaling, basically getting rid of respighi: instead of MODFLOW, we can directly call this function.

[JoerivanEngelen]

Some screenshots for context: The poisson interpolation allows interpolating ditches with recharge.

This would looks something like this for ditches:

Where purple line is the ditch level (boundary condition), blue line is the laplacian interpolation, and red line the poisson interpolation. This is an example for negative recharge AFAIK (was still work in progress and the sign got swapped accidentily, but it was the best screenshot I have to explain the context somewhat).

After fixing the sign errors this turned into:

[sonarqubecloud]

Quality Gate passed

Issues
1 New issue
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud

[Huite]

After some reflection: the respighi scaling method requires us to use "weak" boundaries as well (e.g. RIV or GHB).

With only hard boundaries, if you downscale at 25.0 m, the value at 25.0 m gets burned into the grid directly. This means that if the ditch distance is <50 m or so, you will only get the boundary conditions (like ditch stages). This would mean that for e.g. the peaty areas of South-Holland, the scaling essentially fails.

It's relatively easy to add a robin boundary (a GHB) as optional arguments for the system of equations.

This allows the interpolated head to deviate somewhat from the provided robin values.