There are many facets to depression and to analyse them from a evolutionary perspective, you should evaluate every single one them separately.

Some specific types of depression, like 'burn-out' syndrome, have an obvious cause-effect and is some sort of defense mechanism to protect you from yourself. The metaphor of getting burned is fitting, because it's similar to physical pain.

But evolution doesn't bother if something is beneficial or not. A common mutation can be devastating for the individuals mental health, but as a species we can have evolved to circumvent this weakness by triggering certain behaviors to prevent us from reproducing. The courting behavior in animals is also often a test the behavior of a potential partner; unexpected behavior leads to rejection.

Genes do not evolve into cleaner, more efficient code, but rather into super complex buggy spaghetti-code that somehow works. We may have evolved to be somewhat intelligent, but we are also inherently flawed.

An interesting observation in that regard was the deranged penguin (documentary: Encounters at the End of the World).

https://www.youtube.com/watch?v=x7kdDeGXUjI

To be more precise, it was 40m to simulate 1% of the neocortex.

Using Moores law we can postulate that it takes 17 years to increase computational power a thousand fold and 34 years to increase it a million times. So that should give you more intuïtion of what 1% actually means. In the course of a couple decades it would take 4 minutes to simulate 1 second of an entire neocortext (not the entire brain).

That doesn't sound too impressive either, but bear in mind that human brain <> strong AI. We are talking here about the physics model of the human brain, not the software architecture of an acutal AI. We could make it a million times more efficient if we trim the fat and keep the essence.

Our brains aren't the ultimate authority on intelligence. Computers already are much better at arithmetic, memory and data transmission.

This isn't considered to be intelligence by itself, but amplifies the ability of any AI at a much larger scale. For instance, Watson isn't all that smart because he had to read the entire Wikipedia and a lot of other sources before he could beat people on Jeopardy. But... he did read the entire Wikipedia, which is something no human has ever done.