Computes the natural logarithm of the gamma function evaluated for large numbers using the approximation published in Hart et al, Computer Approximations 1968.
Use the function gammaln().