Yes, I was only focusing on the "physically possible" part. I don't think it makes sense for us to inherently limit our search for such things to the most obvious solution - focus on that first, sure, but don't rule out that non-physically based assumptions are wrong. We can't assume that a civilization capable of producing a Dyson sphere would exactly follow what we assume to make the most sense.

But I can gladly provide some possible reasons:

• A dark forest scenario would mean that you'd want to hide your energy harvesting as much as possible, while also not living close to megastructures that could be discovered from afar
• A civilization focused on exploring surrounding star systems would automatically spread their energy usage around a wide area
• They could be building things they don't want lower civilizations to see (e.g. prime directive), which would necessitate building them in voids (as long as they don't have technology to cloak it)

You're writing as if the assumption of local energy usage is physically given and can't be wrong, but we simply can't know for now. It could be right, or it could be wrong. Again, I agree that it makes sense to assume it to be correct, as it would be a much more easily recognizable marker, but that doesn't mean it's the only option.