Some good discussion! The point about torque in particular. It's conceivable that a remap alters the torque curve in a way to deliver greater peak horsepower, and yet the powerband is narrowed, causing the car to be no faster overall, with diminished drivability. The comparison to sport mode is apt, as that may be exactly what's happening. The throttle response is altered to give the appearance of greater torque, without really having any measurable gain.
Interesting thought about market positioning, it's certainly possible, but you can't remap a N52 3.0i to be a 3.0si unless you upgrade the intake and exhaust as well. I'm confident that the 2.5 is as efficient as possible, given emissions, drivability, durability considerations that were part of the initial design goals. The market position is strictly displacement based in the case of the 2.5, no silly games required.
There's nothing to be found in climate conditions either. Every ECU adjusts it's map internally to accommodate the full range of possible climate conditions. It's only with carburettors where you have to settle for a happy medium. And even if there was some minor benefit to be found at the climatic extreme, most of us live much closer to the middle of climate ranges. Climate adaptations have more to do with cooling capacities and oil clearances than ECU maps.
Without changing the maximum effective air volume, there's little to be done except improve one factor at the expense of other factors. This is why remapping turbos works so well, it's trivial to increase the air volume.