Really put the skin calculations back the way they were this time.  Helps, but the hands of better bodies meshes still end up in the wrong place.
Added transform parameter to ComplexShape::Split because changing the transforms after the fact ruins the skin binds.

Fixed the relative include paths and tried to update solution, but havn't got the precompiled header working right.

Added a max_bones_per_partition parameter to ComplexShape::Split and NiTriBasedGeom::GenHardwareSkinInfo in anticipation of Tazpn adding that capability.

Added != operator to Matrix44.
Added IsSplitMeshProxy function to NiNode.

Added a SetBoneWeights function to NiTriBasedGeom which automatically calculates the correct bone center and radius.
The NiSkinData::SetBoneWeights function now requires the radius and center as arguments.

The code I added to make NiTriBasedGeom select the correct skeleton root was wrong and has been re-written and tested.  It now seems to select the same skeleton root as real NIF files.
The NiTriBasedGeomData::SetUVSetCount function now sets the hasUVs bool value correctly.
Fixed some skin related bugs in ComplexSh::Split.

Created a new ComplexShape class which can currently be used to automatically split the sort of complicated mesh that contains per-point-per-face attributes into NiTriShape classes.  Should support skinning as well, but has not been tested.
Added TexCoord == operator.
Changed type of Triangle members to unsigned short.
Color4 now defaults to all zeros.
NiAVObject::AddProperty now takes a const argument.
Fixed exception in NiTexturingProperty to throw a runtime_error instead of a string.
NiTriBasedGeom::BindSkin should now automatically finds the skeleton root by itself.  The skeleton_root argument has been removed.