Support linearised all-moving control surfaces

[ngoiz]

Minor update to support linearised all-moving control surfaces.

In addition, minor bug fix for surfaces located at T-junctions and for control surfaces located in different beams but with the same control input.

Note, the linear control surface assembly is done on a per-node basis, a feature improvement would be to rework this on a per-element basis.

Minor fixes to the LinDynamicSim solver (useful to visualise the linearised control surfaces)

[ArturoMS13]

Thank you very much for the commit, moving control surfaces give a great flexibility to the solver. In the future it would be great to generate a library with different moving surfaces. It could be similar to the library of Lagrange constraints.

[ngoiz]

Found a bug when having the same control surface (i.e. elevator) split in two beams with different xb directions. This is fixed in e02a388, however, prior to merging the following needs to be done:

• Test implementation in the Horten wing (elevators on the same xb)
• Implement a simple test script

I'm marking the PR as draft until I finish this

[ngoiz]

I have tested the fix for control surfaces that deflect in the same direction for control surfaces on beams with different xbs with positive results.

I included a simple script as a skipped unittest that can be used to visually test changes

[ArturoMS13]

Good to go, thanks for the changes and the code cleaning