The motivation for this is that we sometimes want to access the ir nodes of a ConjectureData that hasn't been frozen/concluded yet, similar to how we sometimes do for the buffer.
Haven't seen that flake before, but I'm unsure how recent changes could have caused it. (and it does look like something that could flake once in a blue moon). Likely unable to investigate until wednesday.
The motivation for this is that we sometimes want to access the ir nodes of a
ConjectureData
that hasn't been frozen/concluded yet, similar to how we sometimes do for the buffer.