Any thought to capturing the forward end of the trailing arms between the shear tabs?
Well I agree and have been thinking about it. One issue is the bodywork gets in the way if the trailing arms are not on the outside. I'll address that with a request for opinions below. Going with what's there in the picture now, I'd say use the long radius rods available in oval track catalogs. That length is sold as "3/4 radius rods" meaning 3/4" rod ends and 1.25" material swedged down for the rod end. This part would use a bolt more then twice as strong as a 1/2" bolt and the rod end itself would weigh about 3.5 times as much.
I should draw that part in just so you can see it. It would probably intimidate the competition. I would also say that single shear is what holds the flywheel on, the clutch too and also the car's wheels. The load is distributed across four mountings, but assume for a moment it's on one mount. Pick 3000 lbs. of force for a round number. Your motor could do that, but I am not sure where the traction would come from. You would be accelerating at 2g's. The off center load would be less the 250 foot lbs. of torque. The compression on the bracket would be about 1/3 the yield strength of a 1" strip of the bracket material. The actual load on the bracket at the "contact patch" of the bolt could be quite high, but I am assuming ( and drawing in ) some support around the bracket holes. Usually a tube welded thru and something like a washer welded on the face.
I am thinking of a bigger change to try and work out what you're saying though. I'll include a picture below of a typical OEM control arm which supports both the weight of the car and also a trailing arm into the mid span of the "wishbone". I see this a lot in cars. It makes the arm heavier, but it may not be so bad for us. The wishbones and radius rods weigh very little compared to the (rear) spindle 10+ lbs., disk 15-20 lbs. wheel 10-20 lbs. tire 10-20 lbs..
What I'm thinking is designing the control arms to have a clevis for the trailing arms. Then we have much more choice in wheel offset because we do not need to attach the trailing arms to the upright. It seems doable to attach to the upright, but it seems the choice is confining us a bit. Once we decide to attach the trailing arms to the upper and lower links then I can make the rear roll hoop just a couple inches narrower then the main frame rails and run the trailing arms straight back between the upper and lower frame rails. The trailing arms fit between the diagonal and the outside skin.
You would need another layer of skin on the inside to protect the passengers from the moving trailing arms.