The rotation idea apealed to the programmer in me but I suspect might look a bit confusing in use. ( I would probably go for the 'shrink the image, use the leftover space for status' approach).
Re single finger/mouse approach. Definitly if this was anything but what it is (and OK I might do it anyway!).
Thanks for taking the time to feedback.
From a development point of view it's of course interesting, indeed from a user side of view it's challenging to make it intuitive. You might use two fingers to control the angle of the magnifying glass, or the dragging direction controls the angle. For
example the glass is always in front of the dragging direction, this could work intuitive enough, and just average the last few coordinates so it's not too jumpy.