Failing landice_FO_GIS_CoupledThickness test

[ikalash]

This test began a couple weeks ago with diffs:

https://sems-cdash-son.sandia.gov/cdash//test/8919018?graph=status

The diffs are tiny actually:

Response Test 0: 1.094731601407e+08 != 1.096541803121e+08 (rel 1.000000000000e-06 abs 1.000000000000e-06)

Sensitivities (0,0):

                    1.829216632090e+07
Sensitivity (0,0), Test 0,0: 1.829216632090e+07 != 1.829143517389e+07 (rel 1.000000000000e-06 abs 1.000000000000e-06)

Presumably these are due to a PR Luca pushed a couple weeks ago.

The diffs are very small. Should we just rebaseline?