While I get your point, an in-flight device that gets stuck in an infinite loop should not cause an in-flight fire. If that still happens, then it's safe to say that more than one mistake was made. Otherwise you're basically assuming that an untampered vendor model of that unit is so perfect that it would never get stuck in a loop. And as we all know in aviation, to assume anything is making an ass out of 'u' and me. Especially when you assume that something is perfect and cannot fail.
Notwithstanding this, I totally get what would happen in the scenario you describe, and I think it's pretty accurate.
Again - like I said, I don't think that's a likely scenario at all. I think this guy did his stuff right and so on.
But there's all kinds of unanticipated failure modes when someone without actual knowledge of a device generates a data structure that mostly kinda sorta works, and the FAA does not do anything halfway. The people who designed it okay it, or you get it re-certified, or else it's not approved (except experimental certification, based on what people are saying).
Notwithstanding this, I totally get what would happen in the scenario you describe, and I think it's pretty accurate.