server : add Anthropic Messages API support (#17570)

* server : add Anthropic Messages API support

* remove -@pytest.mark.slow from tool calling/jinja tests

* server : remove unused code and slow/skip on test_anthropic_vision_base64_with_multimodal_model in test_anthropic_api.py

* server : removed redundant n field logic in anthropic_params_from_json

* server : use single error object instead of error_array in streaming response handler for /v1/chat/completions and use unordered_set instead of set in to_json_anthropic_stream()

* server : refactor Anthropic API to use OAI conversion

* make sure basic test always go first

* clean up

* clean up api key check, add test

---------

Co-authored-by: Xuan Son Nguyen <son@huggingface.co>

tools/server/server-common.h

@@ -294,6 +294,9 @@ json oaicompat_chat_params_parse(
     const oaicompat_parser_options & opt,
     std::vector<raw_buffer> & out_files);
 
+// convert Anthropic Messages API format to OpenAI Chat Completions API format
+json convert_anthropic_to_oai(const json & body);
+
 // TODO: move it to server-task.cpp
 json format_embeddings_response_oaicompat(const json & request, const json & embeddings, bool use_base64 = false);
 
@@ -320,7 +323,10 @@ std::string tokens_to_output_formatted_string(const llama_context * ctx, const l
 
 // format server-sent event (SSE), return the formatted string to send
 // note: if data is a json array, it will be sent as multiple events, one per item
-std::string format_sse(const json & data);
+std::string format_oai_sse(const json & data);
+
+// format Anthropic-style SSE with event types
+std::string format_anthropic_sse(const json & data);
 
 bool is_valid_utf8(const std::string & str);