now DeviceInfo provides full information about device (from cudaDeviceProp)

This reverts commit fba72cb6, reversing
changes made to 02131ffb.

split them into several small files

fixed iterations>1 case in morphological operations
fixed possible access violation in HSV2RGB
fixed the case learningRate==0 in BackgroundSubtractorMOG2

fixed gpu performance sample

used random images in gpu filter tests

added additional tests for gpu filters
fixed gpu features2D tests

updated gpu tests for CornerHarris and CornerMinEigen
moved direct convolution implementation to gpu::filter2D, gpu::convolve now use only DFT-based algorithm (Bug #1639)

removed void integral(const GpuMat& src, GpuMat& sum, GpuMat& sqsum, Stream& stream) - it fails with NPP_NOT_IMPLEMENTED error
updated docs, accuracy and performance tests

now it supports kernel's size up to 32

switched to gpu::remap in opencv_stitching

GPU: updated upsample, downsample functions, added pyrDown, pyrUp, added support of 16S filtering; put spherical warper on GPU (from opencv_stitching)