Embedded Unit

there is a bug in embUnit/stdImpl.c - code assumes unsigned long is max 32
bit long, which is not true in general. C/C++ specification says only that
sizeof(int)<=sizeof(long) - nothing more. if you expect unsigned type,
32-bits-long use uint32_t instead (#include <stdint.h>).

fix for this requires addition of 'stdint.h' header and changing 'unisgned
long' to 'unit32_t' in embUnit/stdImpl.c in functions _xtoa() (line 113)
and stdimpl_itoa() (lines 140 and 142).