Just my experiences to add to the information bucket!!

On my Edge 540 I use reflex aileron. No more than 5 degrees and I need to mix a little down elevator trim with it. It adds a little to the landing speed but not so you would notice. What I found is that it does not cure tip stalling but provides for plenty of warning. Prior to adding the reflex, the tip stall would come on without warning and progress rapidly. After adding reflex, the airplane gets mushy near the stall, just the way I expect it to feel (and then it tip stalls). I set the reflex as a fixed amount in the "approach" flight mode. I use this mode for takeoff also and the reflex makes no difference to the takeoff - the airplane is "skull dragged" into the air by a big YS 4 stroke.

Low speed fin effectiveness plays a part in this too. Yaw stability, or lack of it at low speed, contributes to the wing drop associated with tip stalling. The fin could be too small, or blanketed by the fuselage at high alpha (this happens on scale airliners), or blanketed by wing downwash, also at high alpha.

Another trick, which works with models because of their low Re, is to put a turbulator (trip strip) on the outper wing. This can cause early boundary layer separation and, after re-attachment a more energetic boundary layer which remains attached over more of the wing at high alpha. Whatever!!! It works on gliders so it's worth a try. The positioning of the trip strip is critical so some flight testing is needed. A strip of fine monofilament (6lb fishing line) is all that is required along, say, the outer 1/4 of the wing panel. Start off with it at 30% and move it back 5% at a time until upper air testing shows it works.

For your CG. Try not to consider CG location in terms of a percentage of the chord or other such criteria. Determine the mean aerodynamic chord and aerodynamic centre of the airplane and position your CG according to this. There are plenty of web sites explaining how to do this. For your Sabre, I would suggest 30% MAC as a good starting point.