Simplify interpretation of some derivatives (#201).

[Rodhern]

Proposed minor changes to stability derivatives as per #201. In short the main principles for the changes are:

• Lift and drag forces are opposite to the Z (down) and X (forward) directions; therefore positive derivatives to Cl and Cd are most sensibly presented as negative derivatives of Z and X forces.
• For modest pitch, yaw and roll angles, an increase in yaw angle should not greatly affect the craft's angle of attack (orthonormal set of 'velocity', 'liftVector' and 'sideways').

[ferram4]

No. None of these are supported by any flight dynamics literature that I have seen and just seem intended to introduce confusion. Negative lift coefficients for upwards lift forces and negative drag coefficients for forces that slow down vehicles are completely opposite to everything in aerodynamics. In addition, making assumptions about vector magnitudes that need to be 1, when floating point errors and significant yaw angles can cause the vector magnitude to grow is unacceptable.

I'm closing this, unless you can cite flight dynamics literature to justify your changes.