#27 Improper fix for PR#16489

[Roby Joehanes]

Improper fix for PR#16489 in JDistlib

The following regression test from R failed:
ct3 <- system.time(N <- rhyper(100, 8000, 1e9-8000, 1e6))[1]
table(N)
summary(N)
stopifnot(abs(mean(N) - 8) < 1.5)

JDistlib as of 0.4.2 yielded huge N (in the order of 50000).
Refer to: https://bugs.r-project.org/bugzilla3/show_bug.cgi?id=16489

[Roby Joehanes]

Fixed in 0.4.3