- tests for both SIMD128 / SIMD256
- different dispatched + baseline(SIMD128) intrinsics

Found via `codespell -q 3 -I ../opencv-whitelist.txt --skip="./3rdparty"`

These CMake options are enough:
- `-DWITH_INF_ENGINE=ON`
- `-DIE_PLUGINS_PATH=lib/ubuntu_16.04/intel64`
- `-DENABLE_CXX11=ON` (OpenCV 3.4 only)

CMake 3.9.0+ is required

- requires OpenCV source directory
- requires OpenCV binary directory with built modules and 'python_bindings_generator' target

Use "OPENCV_FFMPEG_USE_FIND_PACKAGE" variable to specify find_package() parameters

* Update Intel's Inference Engine deep learning backend

* Remove cpu_extension dependency

* Update Darknet accuracy tests

* imgcodecs cmake: the option to customize supported formats list (WITH_IMGCODEC_HDR, WITH_IMGCODEC_SUNRASTER, WITH_IMGCODEC_PXM)

* imgcodecs: fixes

- fixed CMake scripts (=OFF doesn't really work)
- restore dropped GDCM block
- added _IMGCODEC_ prefix
- fixed tests
- include PAM format under WITH_IMGCODEC_PXM option

issue 11552

- remove duplicates
- restore "always on" missing entries
- fix FP16 detection on MSVC

via GCC/Clang __POPCNT__ define

Reproducer:
    cmake -DCPU_BASELINE=AVX2 -DCV_DISABLE_OPTIMIZATION=ON ...

  * remove NVidia and Nvidia
  * replace Cuda with CUDA
  * keep the letters for API

CMake generated files is a part of target depends list.
Details: `cmake --help-policy CMP0058`

* Add CPU default extensions loading in IE dnn backend

* Load cpu_extensions for the future Intel's Inference Engine