datetime review part 2 [Update 2]

[Jonathan M Davis]

New version: http://is.gd/gEu9W

I decided to try and take advantage of the new purity rules to make more of the 
functions pure, so more of them are pure now, but between the inability to have 
pure invariants and the majority of Phobos functions being impure, I didn't get 
to make anywhere as many functions pure as I would like to.

I also found a bug in Clock.currStdTime() where it didn't convert it from the 
unix epoch to the epoch that std time and SysTime use, so the current year was 
in the first century A.D.

- Jonathan M Davis