Make kernel results depend each other directly

[mehmetyusufoglu]

This simple PR does not affect the performance of the babelstream benchmark or results.

2 of the 5 kernels of Babelstream-benchmark were not connected to each other, with this change if one of them changed somehow and fails; error is cached in the last result. ( Since we dont check after each kernel run, this is needed to make sure all kernels are connected.)

This PR is an extension of previous PR: #2299
A result of run: ./babelstream --array-size=36000000 --number-runs=100

AcceleratorType:AccGpuCudaRt<1,unsigned int>
NumberOfRuns:100
Precision:double
DataSize(items):36000000
DeviceName:NVIDIA RTX A500 Laptop GPU
WorkDivInit :{gridBlockExtent: (36000), blockThreadExtent: (1000), threadElemExtent: (1)}
WorkDivCopy :{gridBlockExtent: (36000), blockThreadExtent: (1000), threadElemExtent: (1)}
WorkDivMult :{gridBlockExtent: (36000), blockThreadExtent: (1000), threadElemExtent: (1)}
WorkDivAdd  :{gridBlockExtent: (36000), blockThreadExtent: (1000), threadElemExtent: (1)}
WorkDivTriad:{gridBlockExtent: (36000), blockThreadExtent: (1000), threadElemExtent: (1)}
WorkDivDot  :{gridBlockExtent: (256), blockThreadExtent: (1024), threadElemExtent: (1)}
Kernels         Bandwidths(GB/s) MinTime(s) MaxTime(s) AvgTime(s) DataUsage(MB) 
 InitKernel      62.381          0.0092335 0.0092774 0.0092557 576    
 CopyKernel      89.822          0.0064127 0.0065134 0.006469 576    
 MultKernel      89.616          0.0064274 0.0065015 0.0064735 576    
 AddKernel       90.715          0.0095243 0.0095734 0.0095416 864    
 TriadKernel     90.899          0.0095051 0.0096174 0.0095446 864    
 DotKernel       93.371          0.0061689 0.0062328 0.0062011 576