转换器: 修复 Responses 流式 response.completed 缺 usage

- 上游 chat 流 finish_reason 块先于 usage 块, 原实现在 finish 时即发
  response.completed 导致 usage 为空; 改为等 usage 块到达(或 [DONE] 兜底)再发,
  completed 携带 input/output tokens

Co-Authored-By: Claude <noreply@anthropic.com>

server/internal/proxy/convert/stream.go

@@ -414,6 +414,7 @@ type chatToResponses struct {
 	sseState
 	model      string
 	usage      any
+	finishSeen bool
 	done       bool
 }
 
@@ -425,6 +426,7 @@ func (t *chatToResponses) line(line []byte) []byte {
 		return nil
 	}
 	if done {
+		// 流结束兜底：finish 后 usage 未随块到达时在此补发 completed
 		if !t.done {
 			t.done = true
 			return eventLine("response.completed", map[string]any{
@@ -453,6 +455,9 @@ func (t *chatToResponses) line(line []byte) []byte {
 			finish, _ = c0["finish_reason"].(string)
 		}
 	}
+	if finish != "" {
+		t.finishSeen = true
+	}
 	var out [][]byte
 	if role, _ := delta["role"].(string); role == "assistant" {
 		out = append(out, eventLine("response.created", map[string]any{
@@ -465,7 +470,8 @@ func (t *chatToResponses) line(line []byte) []byte {
 			"type": "response.output_text.delta", "delta": content, "item_id": "msg_1", "output_index": 0, "content_index": 0,
 		}))
 	}
-	if finish != "" && !t.done {
+	// 上游 usage 块(choices 为空)通常晚于 finish_reason：此时再发 completed，携带 usage
+	if _, hasUsage := m["usage"]; hasUsage && t.finishSeen && !t.done {
 		t.done = true
 		out = append(out, eventLine("response.completed", map[string]any{
 			"type": "response.completed",