Ported kimi-k2 support from llama.cpp (#609)

Original patch by @gabriellarson:
https://github.com/ggml-org/llama.cpp/pull/14654

Co-authored-by: anikifoss <anikifoss>

include/llama.h

@@ -112,6 +112,7 @@ extern "C" {
         LLAMA_VOCAB_PRE_TYPE_FALCON_E       = 35,
         LLAMA_VOCAB_PRE_TYPE_SEED_CODER     = 36, //llama.cpp lists this as 35
         LLAMA_VOCAB_PRE_TYPE_HUNYUAN        = 37, //llama.cpp lists this as 36
+        LLAMA_VOCAB_PRE_TYPE_KIMI_K2        = 38, //llama.cpp lists this as 37
     };
 
     // note: these values should be synchronized with ggml_rope