Value = Logp ( Number AS Float ) AS Float

Computes Logp(x) = Log(1 + x).

This function is supposed to keep the precision when Number is near zero.