OpenGL2: Add GPU vertex skinning for IQM models

Using GPU vertex skinning is significantly faster than CPU vertex
skinning. Especially since OpenGL2 has to run R_VaoPackNormal() and
R_VaoPackTangent() each vertex each frame which causes CPU vertex
skinning to be significantly slower than OpenGL1 renderer.

code/renderergl2/tr_surface.c

@@ -1313,4 +1313,5 @@ void (*rb_surfaceTable[SF_NUM_SURFACE_TYPES])( void *) = {
 	(void(*)(void*))RB_SurfaceFlare,		// SF_FLARE,
 	(void(*)(void*))RB_SurfaceEntity,		// SF_ENTITY
 	(void(*)(void*))RB_SurfaceVaoMdvMesh,   // SF_VAO_MDVMESH
+	(void(*)(void*))RB_IQMSurfaceAnimVao,   // SF_VAO_IQM
 };