IfcCosine

[Elvira2227]

In this pull request adds new implementation for IfcCosine.
The sturcture of the functions getpointOnCurve and getDirectionOnCurve are similar with representation of IfcSpline.

Note:
There are a mistake in allready merged functions of IfcSine, for instance in the declaration of the template functions getPointOnCurve and getDirectionOfCurve. Previously, the code did not reached this functions and it represents geometry only with cartesianPointList #565 . Now this mistake has been fixed.
Another point corresponds to issue #560. The calculation takes quite a long time.

In the example file cosine-curve_100.0_300_1000_1_Meter were are not 100 point as usual, but 101. Therefore, our implementation does not reach the last point. I think, it is a bug in example file.
Here is a screenshot of this example. And we see that the maximum point, which should be the last point, is just the penultimate point.

ToDo:

• Add contexte to ReleaseNotes
• Add context to SupportedIFCrepresentations

[jschlenger]

Looks good so far.
Did you already work on the issue with the long calculation times?

[pjanck]

lftm

Is there an issue/PR for the unit tests?