So I went away for a week of real work and return, only to find this thread appears to have rewound several weeks to repeat a topic already hung, drawn and quartered - or at least it seemed so.
Here's where I thought we were:
a) All the adults seemed to agree that the 'zoom-climb' was induced by PF action. The stick-trace shows PF NU inputs, that correspond with control surface movements, that correspond with altitude change...
b) The NU inputs were not continuous (at this point) and hence not sufficient to demand a trim input from the THS. However later in the event, PF did pull, long and hard NU, such that the THS moved, as demanded, to off-load what the pilot appeared to desire.
c) That the "auto-flight", protections or any other computer induced skulduggery had anything much to do with what went wrong seems to be absent from all the evidence so far, excepting that at autopilot release the aircraft needed a mild roll correction.
d) We have seen that certain contributors here have an amazing ability to add 2+2 and get -47, in other words theories with little basis in fact, but equally little to dispute them, which in the mind of the author therefore gives them legitimacy... much to the frustration of many others. Just because something is possible does not mean it is likely. The shortest route between two points is the straight-line. While it is true we do not have all the 'dots', those that we do have, are pretty convincing.
Anyway, I'm not sure I'm contributing much here, except to get my mind straight, so I'll shut-up for a bit longer.