Why in diagrams of a stall, the relative "wind" comes in horizontally no matter how the airplane is orientated? Doesn't the engine, which produces the thrust for the "wind", stay parallel to the wing making the diagrams of wind at higher AoAs much more inaccurate because of the direction of thrust?
The diagrams are accurate.¹
The thrust, produced by the engine, is one of the forces acting on the airframe. The others are lift, drag and weight (a.k.a. gravity).
(from How It Flies, chapter 4)
When these forces are in equilibrium, i.e. their (vector) sum is zero, the aircraft flies straight, in whatever direction it does (that is what Newton's first law says: an object remains at rest or continues to move at constant velocity unless acted upon by external force).
When the forces are not in equilibrium, the aircraft accelerates in the direction of their sum (that is what Newton's second law says: the sum of external forces on an object is equal its mass times its acceleration). But direction of acceleration and direction of velocity are independent. If the aircraft is turning, acceleration is to the side while velocity is (approximately) forward. If it is slowing down, acceleration is aft and velocity is still forward.
So thrust and direction of flight don't have to be parallel.
For a more practical argument, imagine an airplane doing the tailslide aerobatic manoeuvre. The nose points up and the engine is running and producing thrust up. But because the thrust is less then the gravity, the aircraft slows down until it stops midair and starts to fall back to the ground, tail first. So for a moment, the relative wind will come from aft while the engine is still producing forward thrust. Only then will the aircraft pitch over to restore normal airflow over the wings and return to fully controlled flight.
Now of course since the lift and drag forces depend on the relative wind. The lift, which is the aerodynamic force perpendicular to the relative wind, increases with angle of attack, which is the angle between the relative wind and the wing planform, until stall angle of attack is reached.
So when the plane is flying level slowly, the nose is pointed up to create sufficient angle of attack. The thrust is thus pointing slightly up, so small part of it balances a bit of weight, reducing the lift requirement. That is why aircraft actually stall at slightly slower speed when the engine is developing full power than when it is stopped.
¹ That is accurate regarding the direction of oncoming relative wind. They are not accurate regarding the flow around and behind the wings.