Add ikawrakow/ik_llama.cpp as a new package with CUDA/Vulkan support,
enabling MTP (Multi-Token Prediction) and IQ4_KS quantization. Wire it
into llama-swap with a new 'ik-qwen3.6-27b-iq4ks-thinking' model config
and 'iq36' alias. Also add a chat template download to the vLLM setup
script and include the binary on lin-va-desktop.
