(~ cool graphs and interesting points ~)
The only time a spec revision resulted in a huge boost in sales was the Game Boy Color. While it was short-lived, it posted remarkable sales. While Nintendo never gave specifics as they never separated GBC sales from sales of older models, both NPD and Famitsu data indicate a substantial portion of Game Boy sales were from the Color. In the U.S., the GBC's share of total GB sales was at least 40% (while there's no official NPD tally for GB sales in just the U.S., based on the U.S.'s average share of Nintendo sales in "the Americas" region, it likely sold on the order of 38-39M units; NPD numbers show about 17.3M Colors sold). In Japan, the GBC's share of GB sales was a much more modest 17.5%, but it was still healthy numbers.
But the GBC had the distinction of releasing nearly a decade after the original model Game Boy did. It probably also helped that it A) had a rather high amount of exclusive games that did not work on older Game Boy models, B) was released right around the same time as Pokemon, and C) was in freakin' color. These factors resulted in gamers, especially in North America, treating it as a new console entirely even though it was technically just an upgraded Game Boy.
If we get a "Switch Pro" within the next year or two, I expect the reaction to it will be more like that of the DSi or New 3DS, being treated as simply another model like the Lite. It will give a solid but relatively short-term boost to sales, but will not serve to reverse the inevitable terminal decline in Switch sales. Eventually, everybody that wants a Switch will get one, and only those that want a slightly better model will get that one. The Switch's "true" successor will probably be a completely new platform, a clean break from the Switch. I think it may stand a good chance of being another hybrid, and if it is, it will be a true next-gen experience and not a mere spec upgrade.
So, in your well thought-out post, you mention two possibilities for the reception of a new revision: either it's received like a GBC (basically as if it was a new platform), or like a New 3DS (just a variant). Let's look more deeply at what differentiates the two outcomes. I would suggest (and kind of agree with your post) that the difference lies in the games. You mentioned that the GBC was big because big games came out for it. You mention Pokemon, I could add the Zelda Oracle games, the DX release of Super Mario, the DX release of Link's awakening. Why would DX versions be needed? It's because the added hardware features are so awesome that they require an upgrade in the original games. If I'm not mistaken, the DX games were playable only playable on game boy color, but with today's backwards compatibility options available in the world of game deployment, and with a bit of creativity, it could have been possible to make the games backwards compatible.
Another point your graphs can't show because they are limited to one platform is the importance of games released on the successor to make it fly (launch). If you were able to mesh the curve of one console with that of its successor(s), you would see that it's the games that cause the upswing. And you are also only counting hardware sales, but your blind fspot in the graphs is software sales. In other words, if the switch had an abnormally long life-cycle with waves of games every 3 years, then the result would reflect not in the hardware sales curve, but on the curve of total software sales. Still, interesting graphs!