calc_forces! and calculate_results use the first wing's spanwise_direction for the panels of every wing #358

calc_forces! (src/solver.jl) and calculate_results (src/body_aerodynamics.jl) take spanwise_direction from body_aero.wings[1] and use it to set the lift, drag and side directions of every panel. On a body whose wings have different span directions, for example a vertical fin next to a horizontal wing, every panel from the second wing on gets the first wing's direction. The forces and moments are then wrong, with no error. calculate_results also computes the projected aspect ratio from wings[1].span alone.

solve! has worked on several wings for a while, and #355 now lets linearize take them too, so the Jacobians inherit the same directions.

What I would do: walk the wings with a running panel offset, as the unrefined-section accumulation in calc_forces! already does, and use each wing's own spanwise_direction for its panels. This changes results only for bodies whose wings' span directions differ. A test: a body of a wing along y and a copy rotated 90° about x. The rotated copy's panel forces should equal the one-wing forces, rotated.

Found in review of #355.