The fictitious force is proportional to omega-squared r, and omega is assumed constant. So doesn't that mean the pseudo-potential is proportional to r^2?

--aggghh! Yes, that was dumb. However, my proof remains just as valid as before, and actually gets simpler since the laplacian of this is simpler (namely, now is a negative constant).