[GPU] use parallel ifstream to boost file reading latency #57380

Job	Run time
Smart_CI	16s
Docker	58s
Build / Build	19m 6s
JAX Layer Tests / JAX Layer Tests	3m 39s
Samples / Samples	6m 10s
Pytorch FX Layer Tests / PyTorch FX Layer Tests	18m 34s
Python unit tests / Python unit tests	7m 12s
OpenVINO tokenizers extension / OpenVINO tokenizers extension	2m 31s
Pytorch Layer Tests / PyTorch Layer Tests	39m 58s
C++ unit tests / C++ unit tests	14m 33s
Python API tests / Python API tests	3m 17s
Debian Packages / Debian Packages	1m 45s
TensorFlow Layer Tests / TensorFlow Layer Tests	15m 17s
ci/gha_overall_status_ubuntu_24	3s
	2h 13m 19s

Provide feedback