* Adding functions rbegin() and rend() functions to matrix class.
This is important to be more standard compliant with C++ and an ever increasing number of people using standard algorithms for better code readability- and maintainability.

The functions are copy pated from their counterparts (even though they should probably call the counterparts but this gave me some troube).
They return iterators using std::reverse_iterators

Follow up of an open feature request:
https://github.com/opencv/opencv/issues/4641

* Fix rbegin() and rend() and provide tests for them

* Removing unnecessary whitespaces

* Adding rbegin and rend to Mat_ class with the right parameters so we don't need to repeat the template argument.
An instantiating cv::Mat_<int> for example can call it's rbegin() function and doesn't need rbegin<int>() with this convience addition.

Follows what is done for forward iterators

* static cast the vector size (return size_t) to an int (that is required for opencv mat constructor)
Co-authored-by: NStefan <stefan.gerl@tum.de>

G-API: New python operations API

* Reimplement test using decorators

* Custom python operation API

* Remove wip status

* python: support Python code in bindings (through loader only)

* cleanup, skip tests for Python 2.x (not supported)

* python 2.x can't skip unittest modules

* Clean up

* Clean up

* Fix segfault python3.9
Co-authored-by: NAlexander Alekhin <alexander.a.alekhin@gmail.com>

* fix inliers in solvePnPRansac

* fix inliers in test_usac

* fix inliers in test_usac

Fixes for Swift troubles

* Remove NS_SWIFT_NAME override for Point, Rect, and Size due to Darwin namespace conflict

* Fix swift_type overrides in objc generator

* Add backwards compatibility Swift typealiases for Point, Rect, Size

* Add disable-swift build option to iOS/macOS builds

* Add import directive to swift source when building with disable-swift
Co-authored-by: NChris Ballinger <cballinger@rightpoint.com>

[G-API] IE backend convert fp32 to fp16

* Support FP16 in IE backend

G-API MTCNN demo hotfix to align overall pipeline accuracy with the reference Python code output.

* MTCNN G-API demo aligned with Python from OMZ

* clean up

* more comments from Maxim are addressed.

* address comment from Dmitry

cv::UMat output/input in VideoCapture/VideoWriter (data stays in GPU memory)

* FFMPEG with UMat input/output

* OpenCL_D3D* context

* fix Linux build

* cosmetic changes

* fix build if USE_AV_HW_CODECS=0

* simplify how child context pointer stored in parent context

* QSV interop with OpenCL on Windows

* detect_msdk.cmake via pkg-config

* fix av_buffer_ref() usage

* revert windows-decode-mfx whitelisting; remove debug msg

* address review comments

* rename property to HW_ACCELERATION_USE_OPENCL

* fix issue with "cl_khr_d3d11_sharing" extension not reported by OpenCL GPU+CPU platform

* core(ocl): add OpenCL stubs for configurations without OpenCL

* videoio(ffmpeg): update #if guards

* Put OpenCL related code under HAVE_OPENCL; simplify reuse of media context from OpenCL context

* videoio(test): skip unsupported tests

- plugins don't support OpenCL/UMat yet
- change handling of *_USE_OPENCL flag

* videoio(ffmpeg): OpenCL dependency

* videoio(ffmpeg): MediaSDK/oneVPL dependency

* cleanup, logging

* cmake: fix handling of 3rdparty interface targets
Co-authored-by: NAlexander Alekhin <alexander.a.alekhin@gmail.com>

* Added PaddlePaddle classification model conversion case

* Modify cv2 import as cv

* Modify documents in dnn_conversion/paddlepaddle

* Modify documents in dnn_conversion/paddlepaddle

this corrects bug #16592 where a Stream is created at
each GpuMat::load(arr,stream) call

a correct solution would have been to add a default to GpuMat::load
but due to circular dependence between Stream and GpuMat, this is not possible
add test_cuda_upload_download_stream to test_cuda.py