Use num_threads clause of #pragma omp parallel instead to avoid calling omp_set_num_threads for the entire application

Removed CMake configuration files from gitignore, since builds should be performed out of source tree.

* change tutorial to use enum name instead of value

* use enum name instead of value

Parallelize Canny with custom gradient (#8694)

* New Canny implementation. Restructuring code in parallelCanny class. Align mag buffer and map.

* Fix warnings.

* Missing SIMD check added.

* Replaced local trailingZeros in contours.cpp. Use alignSize in canny.cpp

* Fix warnings in alignSize and allocate just minimum extra columns.

* Fix another warning in map.create.

* Exchange for loop by do loop to avoid double check at the beginning.
Define extra SIMD CANNY_CHECK to avoid unnecessary continue.

Fix race condition in getDefaultAllocator and setDefaultAllocator interaction / not threadsafe currently

* Correct the existing documented T-API functions to match the doxygen format.

* docs: fix comments style

* T-API documentation: minor formatting changes

replaced SSE2 code with universal intrinsics; improved accuracy of the box filter; it should now be bit-exact