I hadn't planned to post my analysis of the video footage, but since I've been moved to write I'll record a +1 for that part of the post (although I think it's more likely that the pilot grossly overbanked while looking into the turn, rather than starting a roll-and-pull through). It looks to me like a straightforward case of the bank angle being corrected without noticing the descent that had been established, followed by a late and excessive pull when the proximity of the ground became apparent.
I very much doubt there is anything straightforward here, as with most aviation accidents its usually a number of contributory factors. Looking at the footage however something else struck me after reading an earlier comment in this thread about an incident in the past of an un-commanded pitch / trim runaway which forced the crew to eject.
I'm not 100% familiar with the segment of the display that was being flown however I have seen footage of a single Gnat running down the display line inverted, I assume as part of the synchro segment.
If this was the manoeuvre being attempted at the start of the longer footage available then its possible the roll to inverted was started, insufficient forward stick was available to keep the nose up and the bank quickly rolled back off to recover. With the pitch up continuing to increase as the trim continues to run to full limit, the g-load and AOA would rapidly increase until the point the aircraft stalls / flicks with no room to recover.
I'm not trying to do the AAIB's job just merely throwing another scenario out there other than simply mishandling at low level, perhaps those with more knowledge of the Gant’s rather complex trim system could discuss?
Regards
UA