kill_with_prepare_conflicts_during_step_up configures a failpoint, then runs code which hits it, then clears the global log assuming this will clear the failpoint, then waits for the failpoint to be hit again. This doesn't work with the new failpoint helper which isn't based on the global log, so we don't actually wait for the second time the failpoint is hit.

I believe this can be fixed just by moving the configureFailPoint down to where we currently clear the global log.