handle grid type scalar

In https://github.com/eWaterCycle/grpc4bmi/blob/6a05c8ac4421647254a042b85056aa3f2800287f/grpc4bmi/reserve.py#L33 we now handle 'uniform_rectilinear', 'rectilinear', but there is also 'scalar' (https://bmi.readthedocs.io/en/stable/#get-grid-type).

Scalar looks like the right type to use for lumped models.

A lumped model might not even know where (geographically) is is being run.

We should figure out what the `get_grid_*` functions should return for a model with `get_grid_type=='scalar'`.

Read grpc4bmi/reserve.py around line 33 and the BMI get-grid-type documentation first. Trace the existing get_grid_* handling for uniform_rectilinear and rectilinear models, then determine and document the expected results for scalar models, including a way to verify those results with a lumped model.

Written by the indexing model from the issue text.