Added new data types to cv::Mat & UMat (#23865)

* started working on adding 32u, 64u, 64s, bool and 16bf types to OpenCV

* core & imgproc tests seem to pass

* fixed a few compile errors and test failures on macOS x86

* hopefully fixed some compile problems and test failures

* fixed some more warnings and test failures

* trying to fix small deviations in perf_core & perf_imgproc by revering randf_64f to exact version used before

* trying to fix behavior of the new OpenCV with old plugins; there is (quite strong) assumption that video capture would give us frames with depth == CV_8U (0) or CV_16U (2). If depth is > 7 then it means that the plugin is built with the old OpenCV. It needs to be recompiled, of course and then this hack can be removed.

* try to repair the case when target arch does not have FP64 SIMD

* 1. fixed bug in itoa() found by alalek
2. restored ==, !=, > and < univ. intrinsics on ARM32/ARM64.

modules/core/test/test_misc.cpp

@@ -31,12 +31,12 @@ TEST(Core_OutputArrayCreate, _1997)
     ASSERT_NO_THROW(local::create( mat(Rect(Point(), submatSize)), submatSize, mat.type() ));
 }
 
-TEST(Core_SaturateCast, NegativeNotClipped)
+TEST(Core_SaturateCast, NegativesAreClipped)
 {
     double d = -1.0;
     unsigned int val = cv::saturate_cast<unsigned int>(d);
 
-    ASSERT_EQ(0xffffffff, val);
+    ASSERT_EQ(0u, val);
 }
 
 template<typename T, typename U>