I did change that checkbox, and I did see the bit change. Without changing my cabling though, the motor goes unstable and has a tendency to run away. I tried various different parameter changes (Sin/Cos bias word (ECT), making Ixx70 negative, etc) but I didn't come across a solution that would cause the motor to run correctly (but backwards).
I guess I'm not exactly sure how the phasing works with the resolver. I followed the instructions in the GEO Brick User manual, and it gets the motor spinning, but I don't know where to look to change the phasing. It's clearly not all the same variables as I use for the Hall phasing.
At the end of yesterday, I swapped two motor power leads (as well as the CW/CCW bit in the ECT) and that works, but I was looking for a purely electrical way to do it, and would still prefer that method.