llama : Add support for DeepSeek V3 (#11049)

* convert : extend DEEPSEEK2 model architecture to support DeepseekV3ForCausalLM by adding EXPERT_WEIGHTS_NORM and EXPERT_GATING_FUNC model parameters and FFN_EXP_PROBS_B tensor type

* vocab : add DeepSeek V3 pre-tokenizer regexes

* unicode : handle ACCENT_MARK and SYMBOL categories in regex

* llama : add DeepSeek V3 chat template, handle new model parameters and tensor types

---------

Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>

include/llama.h

@@ -105,6 +105,7 @@ extern "C" {
         LLAMA_VOCAB_PRE_TYPE_EXAONE         = 25,
         LLAMA_VOCAB_PRE_TYPE_CHAMELEON      = 26,
         LLAMA_VOCAB_PRE_TYPE_MINERVA        = 27,
+        LLAMA_VOCAB_PRE_TYPE_DEEPSEEK3_LLM  = 28,
     };
 
     enum llama_rope_type {