Here's my speculation: The main purpose of the Rafale's canards is to work in conjunction with the elevons to control the aircraft longitudinally by generating pitching moment. The moment that the canards contribute to the the total aircraft pitching moment is due not only to the lift and moment they generate directly, but also due to their effects on the main wing. At high angles of attack, I would expect the canards to exert more overall longitudinal control influence by changing the lift and pitching moment of the main wing than through the forces and moments they generate directly. The canard scheduling you observed on the Rafale suggests the effect of canard deflection might reverse sign beyond a certain angle of attack. For these reasons, I think it would be rather challenging to determine right mix of canard and elevon deflection to best achieve the desired longitudinal response at high angles of attack (particularly challenging to determine through flight testing). If you don't have a way to feed angle of attack into your control surface deflection mixing, I doubt that you could get predictable response by simply reversing canard deflection beyond a certain stick deflection. Without AOA feedback, I can imagine your jet pitching one direction if you pulled back on the stick slowly, and the other direction if you pulled back quickly (probably not a desirable flying quality). Best of luck! I would look to people with Rafale/Gripen/Mirage experience for how they scheduled their canards as a starting point.