Back again.
The original error has returned: "nvme nvme0: controller is down"
I have isolated the issue to one single docker container, Tdarr, however after scrutinising the logs for that I cant find anything to indicate what's causing the issue.
I've added the power state stuff to the bzroot line, I've tried a different physical drive, I just can't for the life of me figure out why my drives are crapping out so hard 🙃