Alcohol evaporation or Ice spray frosting the driver should work as well,
but if you got a part that does not have such a high power loss then often the IR camera is last hope to see less than 1 °C difference, like a dead short inside a MLCCapacitor
I had those bastards on one or 2 drivers, before I had the IR imager, can be a real pain to find on boards that are not typical drivers that contain only a few

This one was my fault, I secure back side with SMD glue, so they stay in position and don’t move outwards during the reflow,
when I placed the AMC on the board on the pad it moved quite a lot outside and smeared the glue to the whole ground pad,
effectively disconnecting it, first time this happened, but one of the small legs not connected happens sometimes

as I always test my drivers on my Rigol DP811 PSU, so I see with Bistro drivers if all AMC delivers the right current in Mode ALL AMC,
on ramping drivers it’s hard to notice at all