Current behavior

Simulation with nose_to_tail rockets and tail_to_nose rockets were not identical given correct inputs. If you did this with calisto, this difference was really small, but for rockets with more complex parts (individual fins, generic surfaces...) this became more apparent.

New behavior

The problem came from the initialization of the initial spin angle. The initialization was corrected and a test was created to assert equivalence of both coordinate systems.

Breaking change

• No