Hmmm.....
I'm thinking that the sander vibrations do absolutely nothing at all in terms of simulating ultrasonic cleaners.
The ultrasonic cleaners work, IIRC, by creating microscopic cavitation bubbles and when those bubbles form & collapse, they remove a particle of contamination from the substrate (base metal) with them.
A palm sander strapped to the outside of a bucket does no more than a good vigorous stirring would, IMO. I'm thinking that any cleaning activity seen here is strictly because of the liquid cleaner itself, along with the heat, as mentioned. I don't think there's any "ultrasonic" action at all.
Ultrasonic cleaners operate at frequencies between 20,000 ~ 400,000 Hz. A 13,000 rpm sander is operating at 13,000/60 = 217 Hz.
So, I'd call it a "homemade agitator" not a "homemade ultrasonic cleaner".