…-android-arm64-build

* fix core module android arm64 build

* fix core module android build when neon is off

When building for Android ARM platform, cmake with
`-D CV_DISABLE_OPTIMIZATION=ON`, the expected behavior is
not using ARM NEON, using naive computation instead.

This commit fix the un-expected compile error for neon intrinsincs.

…d-zlib-shared-since-ndk19

* fix find zlib.so instead of zlib.a when NDK >= 19

On Android platform, `libopencv_imgcodecs.a` is built, expected to
depend on `libz.so`. However, since Android NDK r19, NDK's `libz.a`
is found instead of `libz.so`, leading to link error
(not found libz.a) on machines without same NDK version & direcotry.

Since Android NDK-r19, toolchain pieces are installed to
`$NDK/toolchains/llvm/prebuilt/<host-tag>/...`, including `libz.so`.
Also installed to old paths (`<NDK>/platforms` and `<NDK>/sysroot`)
in NDK r19, r20, r21, but since NDK 22, old paths are removed.

- https://github.com/android/ndk/wiki/Changelog-r19
- https://github.com/android/ndk/wiki/Changelog-r22

With this commit, `libz.so` can be correctly found in NDK<19 and NDK>=19.

`ZLIB_LIBRARIES` is also simplified as `z`, by appending match (regex)
patterns for new toolchain installation directory's libz.so's paths.

* simplify libz.so match pattern for abbreviation

…erload

…otete-lib

* Reduce if statement as it has the same expression on both sides of '&&'

If statement has the same expression on both sides so this can be reduce

Signed-off-by: Dan Ben Yosef <danbey@gmail.com>

* The if statement is to check width and height

* Properly handle empty centers in findCirclesGrid

* Address alalek comments. Add unit test to validate bugfix

* fix build warnings, remove unrelated comment

* Corrected SSD text graph generation for Identity nodes

* Added minor code corrections

- follows iso c++ guideline C.44
- enables default compiler-created constructors to
  also be noexcept

original commit: 77e26a7

- handled KernelArg, Image2D

* Fixed OCL implementation of pyrlk

If prevPts size is (N, 1) (which is a default layout for converting `vector<Point2f>` to `UMat`) the `prevPts.cols == 1` and optical flow will be calculated for the first point only.
Getting `prevPts.total()` as in line 1048 is the correct way to get points count.

* fixed compilation warning (size_t to int)

Signed-off-by: Sergey Krivohatskiy <s.krivohatskiy@gmail.com>

* Update py_setup_in_ubuntu.markdown