I didn't presume any given speed for knife edge flying. If the plane were in stable knife edge flight, we can treat the CG as a point load and the CP as the fulcrum. Since the rudder is the opposition to that, we can tell the mechanical advantage. These are all estimates, but the idea should be clear enough.
If you are at less than a vertical bank the load on the rudder is less. So if you know the maximum force the rudder can exert at any given speed, the gross weight lets you calc the maximum bank angle that the rudder can counter at that speed. That is, if the rudder force is X, (3.74*X)/GW = sin(maxBank). But for any speed fast enough to achieve knife edge flight, the load on the rudder will be the same (within the accuracy limits of this guesstimation) though as the speed gets higher the deflection decreases.
Your graphic is a stripped down version of what I show on page 36. I understand what you're saying, and that's what it takes to make prop four leave the marks on the tracks. But we don't have a smoking gun that makes the case. We have a picture that can be interpreted as you say, but it's not a slam-dunk. And we have a wealth of thin evidence suggesting the vertical bank was impossible. I'm not ideologically set against the vertical bank, but it doesn't seem to be the best explanation of what we find. I wouldn't have questioned it at all if the CAB had correctly reported which side was up, because that's what made me look again.
As for the wings both breaking at the same station, I take your point. The coincidence is an attention-getter, but I don't take it as proof of anything. The manuals don't give any clue that there would be a weak point there. The spars are continuous, the planks are continuous. So it's enough to make one look more closely at those numbers.