diffusion, the migration of carbon atoms in the iron matrix, happens well below red heat. but /significant/ diffusion, recrystallization, graphite flake/nodule growth, or other phase changes, doesn't. and if you're trying to argue that a brake disk is martensitic, you need to think again.
if your gm disk warps in 15k, you need to consider other factors. for honda, elastic distortion caused by incorrect wheel lug torquing has a huge influence. but if it's the disk alone, things like bad post-casting heat treatments, uneven material thickness, etc. can influence whether a disk stays true at high temperatures. the most likely item is cutting corners on heat treatment and reducing heat soak time.
my money's on incorrect lug torquing.
i was a vehicle mechanic for 5 years and also have a materials degree. is that good enough?