add pow(FBig,FBig)

[ertosns]

approximation of pow(x,y) as exp(yln(x)) is required in my current project, until you implement exact pow(float,float) if planed!

[cmpute]

You can use exp(y*ln(x)) by now. I have indeed planned to implement that, but the signature will be powf, and I will have to make sure that it returns the correct value to the precision limit, which might need some work.

Thanks for the PR but your implementation is incorrect in terms of handling the precision limit. I will close it by now.