This isn't a coherent archetype at all. There's 3 different VP strategies in the example list.
1) The Native Village/Bridge deck is a mega-turn engine deck. It doesn't have the interchangeable parts of a deck-drawing engine, but it still looks to build up its deck and make all the green buys in one fell swoop.
2) Chancellor(or Scavenger)/Stash, The Golden Deck, Counting House/Golem, Tactician/Vault, and Apothecary/Native Village deck are all ways of controlling for the green that enters into your deck that is reliant on a particular interaction between cards. These are engines designed to buy one province a turn forever.
3) Deck Deletion Pin decks don't worry about buying green cards; they force the opponent to a helpless position where the pinner can win however they want to.
These are 3 very different things, and don't belong in the same archetype. You can force (1) and (3) into the engine archetype by changing its definition into one that builds up its deck to the point where it just wins (either by buying enough green or forcing the opponent to be unable to play anymore) then leave (2) as the "combo" archetype. Yes, most deck-drawing engines will not win on the spot, but not all examples of an archetype are pure. For the mentioned combos: The Golden Deck has absolute inevitability for a Province buy; Tactician/Vault, Scavenger/Stash, and Apothecary/Native Village has very high probability for a Province buy; and Chancellor/Stash and Golem/Counting House are more like slogs that manage to "cheat" their way into Province money more than they seem like they should from the individual components.
This makes the breakdown of the pure archetypes much clearer based on their approach to victory:
Engine: ignores green cards until it can win.
Rush: buys cheap VP quickly.
Slog: generates whatever VP it can over a long period of time.
Money: buys Provinces one at a time quickly and simply, often missing the target.
Combo: buys Provinces one at a time in a slightly complicated fashion that nearly guarantees a purchase.
Except that there are many, many other combos out there. Take Hermit/Market Square; you can build it up to snap up all the provinces in one go ... or you can setup to buy say 3 prov and have a stacked deck full of golds for then next 2 prov buys. Is that an "engine", not in my books, the Madman are completely dead until you play the combo and sure doesn't
play like an engine. And that to me is the big point about classification schema, they don't exist to be pretty they exist to inform people how stuff plays out. But even if we follow your taxonomy, we end up with combos that don't fit.
Take Kc/Scheme/Count; you can either buy a province a turn and not care (+3 coin, gain a copper, buy a province) - but it also works better with many other alt-VP. Even in a Prov setup, you still will gain a lot of points from Duchy, often nearly as many from Provs. Gardens - gain at least copper/Garden each turn, many turns gain a copper/Garden/Duchy. Dukes - double Duchy turns or Duchy/Duke turns aren't uncommon, this can handily beat some Prov strats. Fairgrounds is also a strong setup - you can often grab a Fairgrounds/Duchy or Unique Card/Duchy.
Fairgrounds, in general, is a godsend to top deck control decks. Yeah you need to spend some time bulking them up, but Kc/Scheme decks are slow to setup and a "second stack of provinces" is just huge.
Other setups, like say Golem/Scavenger/Herbalist/Pstone. Likewise can do things other than grab a single province. When you first hit the combo, you likely can only buy Duchy/Copper. Soon you can buy Province and if the game goes long enough Province/Estate, Province/Duchy, or Province/Province. Again, this is a very Alt-VP friendly combo with Gardens, Fairgrounds (you have 5 kingdom cards, 4 treasures, 3 VP cards, and curse allowing you to wait till your penultimate turn to break the combo get up to 15 unique cards), Duke (quick double Duchy turns if your opponent goes provinces), and maybe even Feodum (never tried it, but gaining Feodum/Silver every turn works in a long enough game).
Or take the Golden deck - some variants, like say Rats or Fortress, don't buy provinces; Fortress in particular scores massively more points if you skip the Provinces. Others, like Kc/Kc/Explorer/Bish/Prov get their points from trashing multiple golds or other gained cards. Even the simple setup of Bish/Money may not be best served by entering the province race. Take something like Lookout/Bish on a slow-ish board. You setup 2 turns slower than Chap/Bish so your opponent may be able to split the provinces 4/4 and win. However, if they go money, you can often drag out the game by Bishing Gold and buying the last province around turn 20.
Other combos - like Possession/Amb have nothing in common at all with other combos. You trash down your deck, Amb over an Amb if needed, and then Possess your opponent. You then Amb his strongest cards (Possession, Prov, Colony) back to you and destroy his deck. With thin enough decks, enough plays of Possession (Tr/Kc, etc.), or enough opponent draw (e.g. Gov) this is pretty much an auto-win as soon as you pull it off. You may truly have zero draw and you may not have perfect odds of winning once you pull the trigger ... but you have high odds of swinging the game and nerfing your opponent. Take an ultra-lean deck of Kc/Kc/Possession/Amb/Amb/Cr. You send over 3 Ambassadors, get 3 Possession turns, and hopefully you get the first hand with 8 cards. You aren't buying provinces (just stealing them), you opponent might still be able to win (if it takes too long to setup), and you don't have any of the normal engine components (draw, cash, etc.)
And this is the real point of the "Combo" category. You can get points from buying a single province, you can get points from VP chips (e.g. Kc/Kc/Monument/Monument/Trasher), you can get points from Alt-VP, you get points from gaining Colonies (e.g. Nv/Peddler/Expand/Nv), you can get points by stealing from your opponent ... the big thing is that these setups
play weird. You may spend many turns buying nothing (or just cantrips) - that's not engine and that doesn't mean buying provinces; you may grab the entire province pile in a turn or you may not touch it the entire game.
And let's be honest a LOT of combos play closer to engine than to Money. Many combos like trashing, money rarely does. Many combos don't, you know, buy money. Most combos are weaker against discard attacks, unlike money decks. Taxonomy should be functional - how do I play this thing, rather than pretty.
