Optimize `merge` algorithm for data sizes equal or greater then 4M items with SLM cache usage #10974

Job	Run time
documentation	20s
clang-format	15s
CPU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	3m 29s
codespell	14s
FPGA_EMU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	6m 5s
HOST,bknd=omp,cmplr=clang++,macos-latest,std=c++17,cfg=release	5m 11s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	11m 38s
HOST,bknd=tbb,cmplr=icx-cl,windows-latest,std=c++17,cfg=release	18m 6s
HOST,bknd=omp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	10m 11s
HOST,bknd=omp,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	6m 16s
HOST,bknd=tbb,cmplr=cl,windows-latest,std=c++17,cfg=release	16m 22s
HOST,bknd=tbb,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	9m 6s
CPU,bknd=dpcpp,cmplr=icx,windows-latest,std=c++17,cfg=release	6m 27s
HOST,bknd=tbb,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	8m 59s
HOST,bknd=tbb,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	10m 52s
HOST,bknd=omp,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	8m 49s
HOST,bknd=tbb,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	10m 50s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	11m 52s
HOST,bknd=serial,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	5m 8s
CPU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	3m 33s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	17m 47s
HOST,bknd=omp,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	18m 34s
	3h 10m 4s

Provide feedback