ISA 2.07 (aka POWER8) effectively extended the expanding multiply
operation to word types. The altivec intrinsics prior to gcc 8 did
not get the update.

Workaround this deficiency similar to other fixes.

This was exposed by commit 33fb253a
which leverages the int -> dword expanding multiply.

This fixes Issue #15506

* Added mechanism to set/toggle NORMAL window to TOPMOST window.

* highgui: remove unnecessary/legacy code

* Adding all possible data type interactions to the perf tests since some
use SIMD acceleration and others do not.

* Disabling full tests by default.

* Giving proper names, removing magic numbers and sanity checks of new
performance tests for the integral function.

* Giving proper names, making array static.

* fix 'chmod' is not recognized as an internal or external command in Windows

https://github.com/opencv/opencv/issues/14939

* cmake: update PCH command handling

* Wrap unsupported by IE layers as custom layers

* Replace pointers to layers blobs to their shapes

* Enable Faster R-CNN with IE backend on CPU

- error: variables in the local address space can only be declared in the outermost scope of a kernel function

* Convert ImgWarp from SSE SIMD to HAL - 2.8x faster on Power (VSX) and 15% speedup on x86

* Change compile flag from CV_SIMD128 to CV_SIMD128_64F for use of v_float64x2 type

* Changing WarpPerspectiveLine from class functions and dispatching to static functions.

* Re-add dynamic runtime and dispatch execution.

* RRestore SSE4_1 optimizations inside opt_SSE4_1 namespace

* removed private flann documentation
* common tutorial images moved to doc/images
* grouping issues

detected by UBSAN

* Fix IE FullyConnected layer

* Fix MyriadX

* Convert lkpyramid from SSE SIMD to HAL - 90% faster on Power (VSX).

* Replace stores with reduce_sum.  Rework to handle endianess correctly.

* Fix compiler warnings by casting values explicitly to shorts

* Switch to CV_SIMD128 compiler definition.  Unroll loop to 8 elements since we've already loaded the data.