Honestly I’m always a bit (actually, a lot) perplexed by the suggestion that the aim of gap reduction is stall of the flap because stall reduces drag. (as anybody with a bit of knowledge of aero knows that usually effect of stall isn’t exactly reduction of drag, quite the opposite)
IMO it’s possible that aero guys from teams describe it saying that closing the gap and “stalling” the flap there’s drag reduction, just to make it short, even if it’s not a totally correct explanation. Give then the term to journalists and there you are with a complete misunderstanding of the phenomenon.
I would be more inclined to think that the “stall”, more exactly, a bit of separation on the flap, is a collateral, and probably unwanted, effect while the main aim of gap reduction is more likely to modify the characteristic of the mutual interference between the two elements, hence reducing downforce and drag generated by the assembly.
In fact the dimension (and shape) of the “duct” formed by the two airfoils has quite big influence on the characteristics of the assembly; there are a few simple rules of thumb (as for example the area must be decreasing or, for maximum downforce you should have the wake of the main element barely “touching” the edge of the flap b.l. ) but obviously the optimal design depends by many parameters, most important being car speed, and by the objective you are pursuing.
One important thing to always remember if that the point of separation of the boundary layer from the surface depends by the energy of the boundary layer itself (the lower the energy, the sooner the b.l. will separate), and the energy depends by freestream velocity.
Stall is a condition that exists when the separation of the flow from the surface happens so soon that the wing basically doesn’t give lift anymore.
(notice : a symmetrical wing at zero angle of attack doesn’t give lift, but certainly isn’t stalled, the b.l. is attached for the whole chord)
A sometimes used definition is that stall occurs in the moment there’s a change of sign of the derivative of Cl-alfa curve. (ie, increasing the angle of attack you usually get increment of lift up to a point, the stall, after that point you don’t get anymore an increment of lift but a reduction)
Maybe this following diagram helps to understand it qualitatively :
At low speed the main purpose of the gap is to avoid complete stall of the assembly (ie, separation of the b.l. quite soon). In that situation in fact, given the low flow energy, the principal advantage of the main element + flap design is spillage of high energy flow from main element high pressure side to enhance the flow on the flap low pressure side, that increment of energy helps the b.l. on the suction side of the flap to remain attached. Consequently you have to design the gap in the best way for this effect. Closing the gap there would really cause the assembly to stall and generate no downforce and lot of drag.
At high speed on the contrary the energy of the flow is lot higher so it’s very probable that also closing/reducing the gap the b.l. is mostly attached with maybe just a bit of separation near the flap trailing edge, so the advantage of the two elements wing compared with a single one isn’t the spillage, but mainly the mutual interference each element has on the other one. That mutual influence changes the velocity (hence pressure) distribution on the two elements and causes increment of downforce generated. Reducing that mutual interference you have reduction of downforce generated and, if done right, of drag.
For this reason, as I said, I doubt that separation on the flap is the aim of the gap reduction/elimination, actually I’m convinced that ideally they wouldn’t want the flow to separate because drag would be even lower without separation. A bit of separation on the trailing edge is accepted because the configuration allows to cut anyway drag compared with the original configuration (open slot), but that’s not because the flap is stalled, it’s simply because of modification on the mutual interference between the two elements.
This explanation would fit more with my experience/knowledge about multi element wings.