Make it work for 32 bits platform with _FILE_OFFSET_BITS=64

[jief666]

There was a arithmetic problem on 32 bits platform when off_t is 64 bits. Basically, we have to make sure to convert to 64 bits BEFORE the multiplication. Hence things like : ((off_t)extent->startBlock) * blockSize.

[jief666]

After this one, if you're still interested in that project, I can submit few pull requests. I worked on : removing memory leaks removing warning extended attributes sparsebundle, clear or encrypted, support adding fuse support UTF8 encoding As of today, mounting an image with libdmg is the exact same as mounting it on Mac. Test with huge images 300-400TB on Mac and on linux 32 bits.