Re-implement SYCL backend `parallel_for` to improve bandwidth utilization #432

Job	Run time
clang-format	15s
HOST,bknd=tbb,cmplr=icx-cl,windows-latest,std=c++17,cfg=release	17m 46s
CPU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	15m 38s
HOST,bknd=omp,cmplr=clang++,macos-latest,std=c++17,cfg=release	4m 57s
HOST,bknd=tbb,cmplr=cl,windows-latest,std=c++17,cfg=release	16m 52s
FPGA_EMU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	24m 27s
CPU,bknd=dpcpp,cmplr=icx,windows-latest,std=c++17,cfg=release	12m 36s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	11m 51s
HOST,bknd=omp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	10m 18s
HOST,bknd=omp,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	6m 11s
HOST,bknd=tbb,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	9m 46s
HOST,bknd=tbb,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	9m 43s
HOST,bknd=tbb,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	11m 9s
HOST,bknd=omp,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	8m 30s
HOST,bknd=tbb,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	11m 10s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	11m 44s
HOST,bknd=serial,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	5m 9s
CPU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	26m 7s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	17m 28s
HOST,bknd=omp,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	18m 45s
	4h 10m 22s

Provide feedback