Error related to transformTopology() and PETSc SF

[baagaard-usgs]

Describe the bug

PETSc SF appears to be incorrect (but the topology is okay) after calling transformTopology() in parallel for some partitions. When the overlap is created, the topology for the overlap is incorrect.

To Reproduce

PETSc branch: knepley/pylith
PyLith fork: https://github.com/baagaard-usgs/pylith.git
PyLith branch: feature-initialize-phases

Steps to reproduce the behavior.

cd tests/fullscale/cornercases/faults-2d
pylith twoblocks.cfg twoblocks_quad.cfg --nodes=3

Associated with #53