prefix caching on repeated workflows with TensorRT-LLM

[emberquill]

My first guess would be this: On GB200, the easiest win is often to back off the aggressive settings and see whether the problem survives.