Complex InverseCotangent is probably wrong

[furun]

i tested the complex.cs trigonometric functions on a mandelbrot set (z^2 + TRIG(c)), and InverseCotangent seams to be incorrect.

mathnet-numerics/Numerics/Trigonometry.cs, Complex Acot, seams to be correct

[cdrnet]

Thanks for reporting this - would you be ready to provide a test and fix (pull request)?

[furun]

i am not a mathematician and would not break it more, but i can tray to fix it.
please check the fix!