Optimize `merge` algorithm for data sizes equal or greater then 4M items with SLM cache usage #11008

Job	Run time
documentation	16s
clang-format	13s
codespell	14s
CPU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	1s
FPGA_EMU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	1s
HOST,bknd=omp,cmplr=clang++,macos-latest,std=c++17,cfg=release	5m 8s
HOST,bknd=tbb,cmplr=icx-cl,windows-latest,std=c++17,cfg=release	9m 16s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	1s
HOST,bknd=tbb,cmplr=cl,windows-latest,std=c++17,cfg=release	9m 27s
HOST,bknd=omp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	1s
HOST,bknd=omp,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	1s
CPU,bknd=dpcpp,cmplr=icx,windows-latest,std=c++17,cfg=release	9m 16s
HOST,bknd=tbb,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	46s
HOST,bknd=tbb,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	1s
HOST,bknd=tbb,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	18s
HOST,bknd=omp,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	8m 28s
HOST,bknd=tbb,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	8m 43s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	1s
HOST,bknd=serial,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	1s
CPU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	1s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	1s
HOST,bknd=omp,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	1s
	52m 16s

Provide feedback