Here's the answer to my CYT #4. If you missed the original video, go watch it and try to come up with your own answer first.
The sequence of events that I observed was:
1. Copter tips back/left.
2. Betaflight OSD shows summary stats screen.
3. Betaflight OSD shows "Armed"
4. Copter hits the ground upside down. vTX goes to static.
5. I hit the disarm switch about here.
6. Betaflight OSD shows the startup screen.
Conclusions:
This was not an ESC or motor issue like a desync because that could not cause the OSD to do the things it did.
This was not simply a brown-out because the OSD disarmed and rearmed before power cycling. Likely that hitting the ground caused the power cycle, independent of the issue that caused the crash.
OSD going to summary stats indicate that the FC perceived a disarm. Disarm was not caused by me flipping the switch. It was not caused by an FC failsafe, because the controls did not lock up, therefore no stage-1 failsafe occurred. The fact that the copter re-armed indicates that stage-2 failsafe did not occur.
Most likely explanation (IMO) is that the receiver went into failsafe, but the receiver did not go "no pulses". Rather, the receiver went to some fixed value of channel settings that resulted in a disarm. This would be possible since I use switch arming, so changing just a single Aux channel to 1500 would kill the copter.
The receiver in this copter is not my usual X4R. It is a mini SBUS receiver that I scavenged from a broken QX90, so reduced range is consistent with this explanation.