KCalc binary fix

[David Johnson]

I've just committed a fix for 154688 that should resolve the erroneous display 
of negative non-decimal numbers (binary, octal and hex). The behavior was 
different on 32 and 64 bit systems. As my access to 64 bit systems is very 
sporadic, I would like some help testing this fix. If you could test this to 
make sure the behavior is consistant I would appreciate it.

In Hex base, The following keys: "0 - F =", should result in FFFFFFFFFFFFFFF1. 
Similar two's complement results should be obtained with binary and octal. 
Any other banging you can do on non-decimal numbers would be appreciated.

p.s. Currently I'm using two's complement for negative non-decimal numbers. 
Some people have remarked that it is not intuitive. I could change this to 
display minus signs instead, but that doesn't feel right to me. Opinions?

Thanks,

-- 
David Johnson