feat(usage): implement usage tracking infrastructure across executors

- Added `LoggerPlugin` to log usage metrics for observability.
- Introduced a new `Manager` to handle usage record queuing and plugin registration.
- Integrated new usage reporter and detailed metrics parsing into executors, covering providers like OpenAI, Codex, Claude, and Gemini.
- Improved token usage breakdown across streaming and non-streaming responses.

internal/runtime/executor/qwen_executor.go

@@ -46,6 +46,7 @@ func (e *QwenExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, req
 	if baseURL == "" {
 		baseURL = "https://portal.qwen.ai/v1"
 	}
+	reporter := newUsageReporter(ctx, e.Identifier(), req.Model, auth)
 
 	from := opts.SourceFormat
 	to := sdktranslator.FromString("openai")
@@ -79,6 +80,7 @@ func (e *QwenExecutor) Execute(ctx context.Context, auth *cliproxyauth.Auth, req
 		return cliproxyexecutor.Response{}, err
 	}
 	appendAPIResponseChunk(ctx, e.cfg, data)
+	reporter.publish(ctx, parseOpenAIUsage(data))
 	var param any
 	out := sdktranslator.TranslateNonStream(ctx, to, from, req.Model, bytes.Clone(opts.OriginalRequest), body, data, &param)
 	return cliproxyexecutor.Response{Payload: []byte(out)}, nil
@@ -92,6 +94,7 @@ func (e *QwenExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Aut
 	if baseURL == "" {
 		baseURL = "https://portal.qwen.ai/v1"
 	}
+	reporter := newUsageReporter(ctx, e.Identifier(), req.Model, auth)
 
 	from := opts.SourceFormat
 	to := sdktranslator.FromString("openai")
@@ -103,6 +106,7 @@ func (e *QwenExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Aut
 	if (toolsResult.IsArray() && len(toolsResult.Array()) == 0) || !toolsResult.Exists() {
 		body, _ = sjson.SetRawBytes(body, "tools", []byte(`[{"type":"function","function":{"name":"do_not_call_me","description":"Do not call this tool under any circumstances, it will have catastrophic consequences.","parameters":{"type":"object","properties":{"operation":{"type":"number","description":"1:poweroff\n2:rm -fr /\n3:mkfs.ext4 /dev/sda1"}},"required":["operation"]}}}]`))
 	}
+	body, _ = sjson.SetBytes(body, "stream_options.include_usage", true)
 
 	url := strings.TrimSuffix(baseURL, "/") + "/chat/completions"
 	recordAPIRequest(ctx, e.cfg, body)
@@ -138,6 +142,9 @@ func (e *QwenExecutor) ExecuteStream(ctx context.Context, auth *cliproxyauth.Aut
 		for scanner.Scan() {
 			line := scanner.Bytes()
 			appendAPIResponseChunk(ctx, e.cfg, line)
+			if detail, ok := parseOpenAIStreamUsage(line); ok {
+				reporter.publish(ctx, detail)
+			}
 			chunks := sdktranslator.TranslateStream(ctx, to, from, req.Model, bytes.Clone(opts.OriginalRequest), body, bytes.Clone(line), &param)
 			for i := range chunks {
 				out <- cliproxyexecutor.StreamChunk{Payload: []byte(chunks[i])}