Finish correct implementation of rotor cubature integration scheme

[Bartdoekemeijer]

Implementing:

• Multiple number of rotor points from N=1 to N=10, equaling 1 to 100 rotor points in total respectively
• Implement correct weighing terms for integration scheme
• Fix the weighted cubic average method for cubature integration

[Bartdoekemeijer]

Check out the new notebook. Everything should work now, as far as I can see.

With Jensen:

And with GCH:

Looks like excellent convergence by N=3 or N=4 for the new method. What left is now implementing compatibility for the CC and TurbOPark solvers, and further validation. Also, while the solutions now converge much better, the model parameters may require re-tuning since the two methods do not converge to the same values.

[Bartdoekemeijer]

And here's the same example for GCH with a more realistic wind shear value:

The effect on T2 is still really big for a full wake overlap situation. This may be because the square grid method emphasizes the wind speed in the middle section of the rotor, while the cubature integration scheme looks more on the outer edges of the circle: