Add GPU models in seperate folder, only old models are supported and they were replaced to make the CPU interface work.
This solves the undesired consequence of breaking the GPU interface.

use ASSERT_FLOAT_EQ to compare float values, it is more robust for
large values

It's needed for std::max.

- updated existing Doxyfile.in
- added corresponding cmake instructions
- added some specific files (layout, icon)
- clean existing doxygen warnings

Conflicts:
	CMakeLists.txt
	doc/CMakeLists.txt
	modules/core/include/opencv2/core.hpp
	modules/core/include/opencv2/core/base.hpp
	modules/core/include/opencv2/core/cuda.inl.hpp
	modules/core/include/opencv2/core/mat.hpp
	modules/core/include/opencv2/core/matx.hpp
	modules/core/include/opencv2/core/types.hpp
	modules/flann/include/opencv2/flann/lsh_table.h
	modules/imgproc/include/opencv2/imgproc.hpp

In CUDA 6.0 there was a bug in NPP LUT implementation (invalid results when
src == 255). In CUDA 6.5 the bug was fixed.

Replaced NPP LUT call with own implementation (ported from master branch) 
to be independant from CUDA Toolkit version.

see https://github.com/Itseez/opencv/pull/3137

add missing stream parameter to call_resize_linear_glob

it uses float arithmetics

Fixed an issue with weight assignment causing the resulting GMM weights to be unsorted in the CUDA and OCL versions of BackgroundSubtractorMOG2

add missing __syncthreads to edgesHysteresisLocalKernel

replace std::swap with own code

* BruteForceNonLocalMeans
* OpticalFlowBM

* out of border access in edgesHysteresisLocalKernel
* incorrect usage of atomicAdd

This reverts commit dbce9069.