linearize and make_dual_shadow accept a BodyAerodynamics with more...

[1-Bort-1]

linearize and make_dual_shadow accept a BodyAerodynamics with more than one wing

Split out of #330 (stability derivatives and trim search on linearize), which needs it to cover multi-wing bodies.

linearize (src/solver.jl, the length(body_aero.wings) == 1 guard) and make_dual_shadow (its own one-wing guard) reject a body with two wings. The va and omega columns need nothing per wing; theta_idxs/delta_idxs are one entry per unrefined section of the one wing, so for several wings they would run over the unrefined sections of all wings in order, matching moment_unrefined_dist, which the solver already sizes by the sum over wings.

Test: the two-wing bodies in test/body_aerodynamics/test_body_aerodynamics.jl (solve! on a two-wing body), Jacobian against AutoFiniteDiff as test/solver/test_forwarddiff.jl does for one wing.

[1-Bort-1]

Lift the one-wing guards in linearize and make_dual_shadow: the shadow rebuilds every wing, and theta_idxs/delta_idxs run over the unrefined sections of all wings in order, split per wing for unrefined_deform!. Test: Jacobian of the two-wing body against AutoFiniteDiff. Built on main, not stacked on #340: it touches one line of make_dual_shadow (the Solver call), a few lines from mine. Leaving the other one-wing guard (body_aerodynamics.jl:469) alone unless linearize reaches it.