Remove unnecessary kercn limitation of 4.

When accessing global memory by DWORD4, memory bandwidth
can be fully utilized on Intel platform. This patch will
make more image format(e.g. 8UC4) be processed in DWORD4
by work-item. After applying this patch, 3 subcase of
./opencv_perf_core --gtest_filter=OCL_RepeatFixture_Repeat.Repeat/*
can be speedup on HD4000 graphics card with Beignet:
OCL_RepeatFixture_Repeat.Repeat/2, 64% improvement.
OCL_RepeatFixture_Repeat.Repeat/6, 50% improvement.
OCL_RepeatFixture_Repeat.Repeat/8, 56% improvement.
Signed-off-by: Chuanbo Weng <chuanbo.weng@intel.com>

modules/core/src/copy.cpp

@@ -846,7 +846,7 @@ static bool ocl_repeat(InputArray _src, int ny, int nx, OutputArray _dst)
 
     int type = _src.type(), depth = CV_MAT_DEPTH(type), cn = CV_MAT_CN(type),
             rowsPerWI = ocl::Device::getDefault().isIntel() ? 4 : 1,
-            kercn = std::min(ocl::predictOptimalVectorWidth(_src, _dst), 4);
+            kercn = ocl::predictOptimalVectorWidth(_src, _dst);
 
     ocl::Kernel k("repeat", ocl::core::repeat_oclsrc,
                   format("-D T=%s -D nx=%d -D ny=%d -D rowsPerWI=%d -D cn=%d",