Development to support the slab test

[whlipscomb]

This PR includes code updates to support the slab test in directory ..tests/slab. The test case was proposed by Dukowicz (2012, doi:10.5194/tc-6-21-2012). CISM applications are described in Robinson et al. (2022, doi:10.5194/tc-16-689-2022). The test consists of a uniform, infinitely long ice sheet (a 'slab') moving down an inclined plane by a combination of sliding and shearing. Analytic solutions exist and can be compared to the CISM solution. The test is useful for comparing different Stokes approximations (B-P, DIVA, L1L2, SSA, SIA, etc.) in different physical situations (e.g., different values of ice thickness, bed slopes, Glen's 'n', viscosity, and basal friction) at different resolutions and time steps. Robinson et al. (2022) used the test to evaluate the stability of different solvers at a wide range of resolutions.

The slab directory contains python scripts to run the test and plot results. Matt Hoffman wrote the original version, which had limited functionality. With this set of commits, the test is fully supported.

These commits includes the following code mods and upgrades:

• python3 compatibility for the slab test
• a new script 'stabilitySlab.py' that runs the stability tests described in Robinson et al. (2022)
• changes to make the L1L2 solver more accurate and robust
• a hybrid solver option that SIA and SSA solutions
• made Glen's 'n' a config parameter (called n_glen) instead of a hardwired constant; it is now real(dp) instead of integer

One known limitation: The plotting script produces two plots. The first shows velocity profiles and works fine, but the second has an issue with displaying results along the y-axis.

[gunterl]

The changes to python 3 work as attended. Performed tests (outside of the new slab test) and they are BFB with those on main. The slab test runs as intended.