[patch] support 14bit NEF

[Hubert Figuiere]

> That assert was in line with the idea of testing for bugs and not
> invalid file format. In particular, it found a bug on my
> implementation: I was skipping one byte in every 10 instead of one
> byte in every 10 columns.
> 
> With the attached patch I was able to decode the D100 file. I also
> changed the function signature a bit since the outsize was not being
> used.

The good news is that this test pass.
The bad news is that it fail right after:

running test NEF-test D1 on file /home/hub/samples/nef/D1/DSC_0489.nef
testsuite: ../../lib/unpack.cpp:67: size_t
OpenRaw::Internals::Unpack::unpack_be12to16(uint8_t*, const uint8_t*,
size_t): Assertion `size % (15 + pad) == 0' failed.
/bin/sh: line 4: 28942 Aborted                 (core dumped) ${dir}$tst

You can find the corresponding file at:
http://www.steves-digicams.com/nikond1/samples/DSC_0489.nef


> > [1] I actually started working on the missing bit to allow running the
> > test suite anywhere.
> 
> I was able to use it  by creating a /home/hub :-)

Actually the idea is to allow downloading those file I know where to
find from the internet, to use them locally.


Hub