Anyway I wonder if it would be possible that the ride forgot the train was sat at the top of the lift hill and began to move the chain back into position for the next train causing the current train to go over. If im correct the smiler lift hill chain has only a single point at which the car can connect, therefore it will always need to continue going round at some point to return this "catch" point to the bottom of the lift hill. Could this have had anything to do with it? Just my thoughts.
Personally I'd have the control system monitor the current drawn by the lift hill motor so that it can tell if it's running, stalled, stopped or lifting a train. That would be fairly trivial to implement, but there's no guarantee that it was set up to do that.
I'm pretty certain that at the very least that the ride system is aware that the lift hill is engaged with a car and is therefore pulling it
I imagine that the main point of investigation for the HSE will be to find the cause and reason as to why, after the ride system performed a safety stop and flagged an error, did the lift hill engage and allow the train to leave the crest of the hill. They'll work backwards from there using root causes analysis until they have no further questions. It's clear that the issue and scenario making all parties involved uneasy is the first lift hill, which will explain the repeated cycling of the ride and stoppages at that point today.