AeroDyn: Improved coning and prebend modeling in future BEM

[ebranlard]

This pull request ready to be merged.

Feature or improvement description
This pull request:

• includes features of the new BEM for large coning and prebend. These features have now been tested (see plots below) and reported (official documentation to follow).
• uses double precision for the calculation of induction factors in the BEM code
• further use the internal AeroDyn flag BEM_Mod to switch between BEM implementations
• reduces the differences between the legacy AeroDyn15 BEM code and the future BEM code that is still being tested and implemented.
• exposes the new BEM interface via the temporary flag WakeMod=10, 11, 12 (similar to 0, 1, 2 but with the new BEM), to allow users to experiment with this feature. WakeMod=10,11,12 will be removed in the future.
• fixes an issue with OLAF when WrVTK=2 (should output VTks only at init and end of simulation)

Documentation, test cases, and support for large yaw angles will be provided in future pull request (handled by another project).

Impacted areas of the software
AeroDyn

Test results, if applicable
No change in the r-test results

Verifcation results
Test results for the IEA 15MW (straight blade) at 9.0m/s and 6.4 RPM, use WakeMod1=1(legacy BEM), 11 (future BEM) or 3 (OLAF).

Typical input files
Simulations, IEA 15MW (straight blade) at 9.0m/s and 6.4 RPM, 50deg coning:
ConeDriver.zip

Acknowledgments
The work was developped as part of the BAR project (@ptrbortolotti ) and in collaboration with Envision Energy (@bjonkman).