fix(ai): use Gemini streaming endpoint in connection test

crates/dbx-core/src/ai.rs

@@ -1116,7 +1116,9 @@ pub async fn test_connection_core(config: &AiConfig) -> Result<AiTestConnectionR
             res.bytes_stream()
         }
         AiProvider::Gemini => {
-            let ep = resolve_endpoint(config);
+            // Gemini only returns SSE from streamGenerateContent; generateContent
+            // returns one JSON document even when alt=sse is supplied.
+            let ep = resolve_gemini_stream_endpoint(config);
             let res = client
                 .post(&ep)
                 .header(CONTENT_TYPE, "application/json")
@@ -2479,11 +2481,11 @@ mod tests {
         apply_chat_completion_thinking_toggle, build_ai_http_client, build_responses_input_with_tools, claude_headers,
         claude_system_prompt, drain_next_stream_line, emit_responses_function_call_item, gemini_text, is_kimi_model,
         openai_response_text, openai_stream_reasoning, openai_stream_text, parse_model_list_response, resolve_endpoint,
-        resolve_model_list_endpoint, responses_function_tool, responses_max_output_tokens, responses_stream_text,
+        resolve_gemini_stream_endpoint, resolve_model_list_endpoint, responses_function_tool,
-        responses_text, responses_token_usage, set_chat_completion_token_limit, stream_data_payload,
+        responses_max_output_tokens, responses_stream_text, responses_text, responses_token_usage,
-        uses_anthropic_messages_api, validate_config, AiApiStyle, AiAuthMethod, AiConfig, AiMessage, AiModelInfo,
+        set_chat_completion_token_limit, stream_data_payload, uses_anthropic_messages_api, validate_config, AiApiStyle,
-        AiProvider, AiReasoningLevel, StreamToolEvent, StreamingToolCallAccumulator, ToolCallRef, AUTHORIZATION,
+        AiAuthMethod, AiConfig, AiMessage, AiModelInfo, AiProvider, AiReasoningLevel, StreamToolEvent,
-        CLAUDE_DEFAULT_SYSTEM, TEST_PROMPT,
+        StreamingToolCallAccumulator, ToolCallRef, AUTHORIZATION, CLAUDE_DEFAULT_SYSTEM, TEST_PROMPT,
     };
 
     /// Reproduce the "Unknown tool:" bug: some OpenAI-compatible providers
@@ -2644,6 +2646,10 @@ mod tests {
             resolve_endpoint(&gemini),
             "https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-pro:generateContent"
         );
+        assert_eq!(
+            resolve_gemini_stream_endpoint(&gemini),
+            "https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-pro:streamGenerateContent"
+        );
 
         let ollama = AiConfig {
             provider: AiProvider::Ollama,