llama : refactor get / set state + remove redundant kv cache API

[ggerganov]

• Normalize the code style
• Move the definitions at the correct place in llama.cpp
• Retire llama_get_kv_cache(), llama_get_kv_cache_size() and llama_set_kv_cache()

Not sure how to test this - maybe we need to add an example, or extend main with store/load state functionality