This whole thread has been fun to read through, and the whole experiment really showcases how much of a geometry problem aerodynamics are.
Vertical stabs are such simple solutions to flight; they provide an opposing force left or right to resist diverting flight. Typically they get larger with respect to the wing area since wings themselves cause yawing forces when in flight and depending on their orientation.
replacing the fin as you've already shown is tricky, cause you have to figure out where else to get your sideways stabilizing forces, and minimize other areas that can cause stability. Swept wings are one of the most effective ways to get sideslip stability, because by nature of being at an angle to forward flight they essentially point some of their lift force sideways. Any slip of some degrees point more lift of whatever wing out sideways and the opposite wing more forward, causing a yawing force to point that noise back straight.
That phenomena is likely why you see more instability with dihedral on your swept wing designs than anhedral. That dihedral is pointing lift back inwards, reducing your stabilizing yawing forces. Anhedral and also the extreme washout points more lift sideways.
On the sizes you're testing those geometries will struggle to balance out, but they could run more effective with larger wingspans. I think one other already suggested such on this thread but that's basically how Prandlt flying wings work. Otherwise you will have to replace your vertical fin with drag rudders, differential thrust, and a whole matter of other tricks.
