seems like you have your answer but if you have a mixed shop and these are the only ones you have issues with then yeah it is the switch maker that is just trying to push off the problem.
I have a full cisco farm and switches from 12 to 2000 ports and have topology turned on for all of them.
Lean on the switch maker to fix the issue.
Thought I'd ask what, if any, feedback you got from Solarwinds around this issue (you mention you were going to open a support ticket?)
Out of interest, were you seeing the high CPU usage through the standard CPU monitoring that's " out of the box " or did you have custom pollers in place akin to this thread Extreme Switch in NPM unable to show cpu and memory problems ?