Optimize `merge` algorithm for data sizes equal or greater then 4M items with SLM cache usage #11005

Job	Run time
clang-format	10s
codespell	18s
documentation	17s
CPU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	9m 22s
HOST,bknd=omp,cmplr=clang++,macos-latest,std=c++17,cfg=release	4m 16s
FPGA_EMU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	13m 15s
HOST,bknd=tbb,cmplr=icx-cl,windows-latest,std=c++17,cfg=release	8m 37s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	9m 39s
HOST,bknd=tbb,cmplr=cl,windows-latest,std=c++17,cfg=release	16m 4s
HOST,bknd=omp,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	9m 6s
CPU,bknd=dpcpp,cmplr=icx,windows-latest,std=c++17,cfg=release	3m 19s
HOST,bknd=omp,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	6m 10s
HOST,bknd=tbb,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	8m 54s
HOST,bknd=tbb,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	9m 18s
HOST,bknd=tbb,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	10m 58s
HOST,bknd=omp,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	9m 4s
HOST,bknd=tbb,cmplr=clang++,ubuntu-latest,std=с++17,cfg=release	10m 58s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++17,cfg=release	8m 55s
HOST,bknd=serial,cmplr=g++,ubuntu-latest,std=с++17,cfg=release	4m 58s
CPU,bknd=dpcpp,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	7m 58s
HOST,bknd=tbb,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	8m 27s
HOST,bknd=omp,cmplr=icpx,ubuntu-latest,std=с++20,cfg=release	8m 14s
	2h 48m 17s

Provide feedback