Low-power llama.cpp server built on Intel N100 and RTX 5060 Ti

Build pairs an Intel N100 mini-ITX board (CW-NAS-ADLN-K, DDR5, 6x SATA, 2x NVMe) with an RTX 5060 Ti to run llama.cpp inference. The project started after the author's old ASRock J1900 system died.