We did evolve smaller eyes when we transitioned to being diurnal hunters.

https://www.sciencedirect.com/science/article/abs/pii/S0047248406002053

One thing you may be missing which is probably relevant here is that, all other things being equal, a smaller eyeball means a larger depth of field. That is, more of the visual field measured along the axis of sight will be in approximate focus the smaller the eyeball. This might sound like a desirable property, and in some cases it is. But it reduces your ability to determine target distance through adjusting the optical properties of the eye, for example by flexing the lens or adjusting the pupil diameter. When you focus on something in the nearfield or midfield, with a smaller depth of field you get more information about the relative position of a target and its visual background than you do with a larger depth of field. In the case of a pinhole aperture which has, in theory, unlimited depth of field, you get no information at all about the distance to any of the objects in the scene based on their relative sharpness because everything has the same sharpness.

For diurnal raptors, for example, a large depth of field isn't much of a disadvantage because they're trying to acquire relatively small targets which are relatively far away and are pretty much always just about the same distance away as their visual background. If you're looking at a rabbit from above you don't need to have optical information to know that the rabbit is just about as far away as the ground. On the other hand, if you are a diurnal ape operating along the surface of the Earth looking at the horizon, and your typical hunting targets are at most single to double-digit meters away when you begin your attack, and which is often hunting targets with complicated backgrounds at different distances away, it's pretty important to have the ability to distinguish how far away your target is in an absolute sense, as well as how far it is from its background. A larger eyeball, which, all other things being equal, gives you a smaller depth of field, helps in that aspect.

In general, our brain does quite sophisticated visual processing on the information it receives from our eyeballs. [In fact, the image processing is so sophisticated that it allows us to surpass the physics of our eyes in some sense under certain conditions.] (https://en.wikipedia.org/wiki/Hyperacuity_(scientific_term)) Obviously we are not actually breaking physics with our eyes, but taking advantage of the fact that we have a matrix of receptors and combining the signal from more than just one to provide information at a finer resolution than would be possible if one simply naively took the intensity values recorded by each receptor and displayed them without processing. One could argue that perhaps shrinking the eyeball would free up some space in the cranium for more brain, but in order to maintain the same visual abilities one would probably need to devote more of the brain to visual processing.

One thing you are kind of approaching here is the fact that evolution has no "purpose" other than increasing fitness with the tools it has available (previous adaptations and random mutations).

It's possible that bird eyes would be more advantageous than ours, but we diverged from them long ago and going back would be pretty difficult with potentially no useful steps in between