A 110 story skyscraper is designed to stand straight up. It is not designed to hold together if the whole building is tipped over at an angle - this doesn’t mean that engineers sit down and say “How can we design this building so that it will not withstand the following load”, it means that said loading condition is not part of the design criteria in the first place. Designing something the size of WTC1/2 that will stay in one piece while leaning at, say, a 45 degree angle, would produce a very different building than existed.
Tall buildings are certainly designed with lateral loads in mind (wind and earthquakes), but those are a fraction of the dead load of the structure itself.
If you could somehow start tipping a structure that size over without snapping it, you’d very quickly hit a point where its own weight would overwhelm the lateral strength and the whole thing would start to fall to pieces. As soon as that happens all those chunks are going to start going straight down.
To give a very quick overview of how a collapse like WTC takes place, consider that you’ve got 110 floors of enormous area each, stacked on top of each other. It’s important to keep in mind the massive scale we’re dealing with, btw - it’s not like a garden shed that can hold itself up if you flip it on its side.
There’s a massive impact in one area which knocks out many of the vertical supports, locally. All the load which was being held by those supports now gets transferred to whatever alternate paths exist - so now other parts of the structure are getting overloaded. Something to think about is that a typical load/resistance combination is 1.4dead + 1.7live = 0.9*resistance, that’s the kind of limiting case that you deal with in structural design. Now in the remaining columns we’re probably carrying considerably more dead load than designed for, even given that 1.4 safety factor, but perhaps the live load isn’t too high at the moment so things are holding up.
Now we’ve got the fire - a few tons of jet fuel plus all the flammable stuff found in any office building. When structural steel gets heated to something like 800F it will lose about 50% of its strength (I can look up the exact numbers if you like). So now your already overloaded structure is facing the loss of half its strength - again, far worse than the 0.9 reduction used in the original design. Not surprising that something is going to give. This means that more vertical support is lost, and all that load has to go somewhere - you can see what’s going to happen, you get a progressive collapse where the weakest, most overloaded areas fail and all that load is dumped onto whatever’s handy, which in turn fails, reducing available support still further, and so on.
We haven’t even considered the fact that we aren’t dealing with a nice, gradual transfer of load - we may be looking at a sudden, catastrophic loss of vertical support so that one floor pretty much vanishes, and the entire dead load of all those floors above it drops down not in minutes or hours but in a second or less (ever stand on a coke can and then tap the side with your other foot - bang, the can is suddenly a pancake in an instant) - multiplying the damage to whatever is beneath. If I carefully hand you a 60 pound bag of sand you can hold it, if I drop it ten feet and tell you to catch it what happens?
So all those (relatively undamaged) floors above the impact/fire zone crash down, destroying the floor underneath the impact/fire zone. That whole mess just keeps going, with the mass of rubble increasing the load on each remaining floor it encounters.
Now recall that we’ve got enormous floors - hundreds of feet on a side. It’s not a single solid object, it’s a crumbling mass of overloaded steel and concrete. Sure some percentage will get scattered about, but the vast majority of it is going straight down and you’ll get a big heap of debris at the bottom.
