Excellent analysis and new exclusion of DeBroglie/Bohm & t'Hooft. Clearly & rigorously argued. My one concern was anticipated in your brilliant last paragraph, but that does then leave a THIRD option not addressed. I alluded to it above, with a proof in my last years essay, with confirmation code & plot in Trails essay. It suggests the elusive 'state after collision'. Your unique understanding, and ability to see QM may be emergent, should allow you to assess it;;
I propose that your conclusion; Randomness "contradicts" determinism needs better definition. My questions on spinning sphere momenta ('OAM') above use a simply 'collision' momentum transfer analogy. A random axis 'pair' state meets Bob's electron, & TWO part vector addition means A,B outcomes are independent!
For collisions on the equator, amplitudes for the question; "clockwise or anticlockwise", will be deterministic, but ALSO uncertain (50:50) so 'random'.
For collisions exactly at a pole; amplitudes for left/right? will be similarly uncertain, but also still deterministic in classical mechanics terms.!What I pointed out in Q5 was that the momentum CHANGE RATE (as Bob turns his dial) across the surface between 0 and 1 is by the cosine of the 'angle of latitude' of the tangent point!. (I then derive Cos2Theta.)
I suggest this is a very important finding, going way beyond just your essay (though meaning your conclusion needs a little tweaking), but ALSO that your last paragraph is correct, as it emerges from a 'Higgs condensate' gravity model-see my essay) and a Gell-Mann 'Quasi' classical derivation of QM should exist, though it may need your help to complete!
If you have a sec do also see my top peer scored 2015 'Red/Green sock trick' essay.
Thought I'm suggesting a minor 'incompleteness' in your analysis, of course agreement on content isn't a valid scoring matter. Well done, and thanks.