llama.cpp adds Ling 3.0 VL support for local inference

The llama.cpp repository now supports Ling 3.0 Vision-Language models. This update enables local execution of Ling 3.0 architectures within the ggml inference runtime.