I would be very interested to see images of the modifications you and Greg Bailey made. You and he appear to share our concern for driver safety.
Your photo validates my point: Your use of stronger steel in the mounts than the control arms, and isolating them from the driver's compartment resulted in the energry transfer from the impact to be contained within the suspension pieces, causing the steel control arms to bend. The proposed solution of bolting the control arm mount to the aluminum seat back would have transferred that same energy load directly to the driver's back, instead. The flat aluminum plate of the seat back would fail before the steel control arms would have bent. In this scenario, the control arm would not have needed to snap and penetrate the cockpit to injure the driver. The amount of distortion we can see in your steel suspension components would have been the minimum amount of movement of the seat back panel into the driver at a very high and concentrated G-load caused by the impact.
Testimonials, discussions of thicknesses and robustness of materials, and comparisons to 49 year-old designs don't change the fact that this particular combination of design, materials and their installation (use of welding) failed when in real world use, at low G forces, without the presence of an impact.
Upon examination after the failure, we came to the same conclusion Craig Cumming and Greg Bailey: that these components of the chassis need a complete redesign to be both strong and safe. If anything, Craig's shunt is a testimony to the integrity of their redesign work. Even under the force of a severe impact, their redesigned mount didn't fail and the driver's compartment wasn't compromised. Well done. We are looking forward to seeing your work.