This is not a direct answer to your question because I don't think there are clear authoritative answers to it given the way you pose it.

Typologies are to a significant extent what you make of them: they are imaginary constructs that we use to order our analysis of a large number of cases. Which means unless you're dealing with a typology that is highly regulated by a specific institutional group (say, the names of species) where compliance with established norms is necessary for scholarly work, you're free to propose, modify, or ignore typological distinctions as you find it useful to do so. There is not a strict right or wrong answer in using typologies of most kinds, and drawn-out arguments about typological precision tend to be deeply unedifying in most kinds of historical analysis.

So take your hypothetical: I think you can classify it however you like. Your scenario is sufficiently complex in one sense (and yet not nearly as messy as real-life warfare often is: the example feels more like something from a computer game) that I think you could sum it up analytically however you like: a loss for everyone, a victory for everyone, a draw. More importantly, in military history questions of whether a particular engagement is a draw or a victory are never purely tactical: they are also always strategic. Is the leader important? Is the city important (just being big isn't enough of a measure of that)?

Moreover, the hypothetical scenario you have is so weird in its specifics (it also reads like a plot for a fantasy novel), it doesn't seem profitable to spend energy trying to classify it. An entire army gets in a city through subterfuge (nod to the Iliad here), kills the enemy ruler through assassination (what, do they all just hide inside of the gate while a lone ninja in the army sneaks off to the house of the city ruler?), then kills half the population of the city while also fighting the enemy forces but retreats just because their own leader is wounded. That's a hell of a cult of personality--an army in battle in a city, where their formations would have to be somewhat dispersed, simultaneously engaged in genocide, retreating suddenly though in an orderly fashion just because their general is wounded. And yet somehow that retreating army destroys city walls on retreating that they weren't able to destroy from the other side (hence the subterfuge)--it's not as if city walls are easier to destroy from the inside as the outside. No wonder you can't think of a battle that fits the hypothetical--the scenario doesn't really make sense, not the least because you're not really thinking clearly about what battles inside of walled cities were actually like.

But even if it did, don't spend too much energy trying to classify cases unless there's some real value to doing so, and unless you're dealing with a very large array of cases that you want to compare for some reason.