feat: LLM 调用日志 + ModelSelector 优化 + devtools.bat 编码修复

- 新增 call_log.go: 全局环形缓冲区记录每次 LLM 调用(模型/Token/耗时/错误)
- OpenAIProvider.doChat/ChatStreamWithTools 自动记录调用日志
- ai-core 暴露 GET /api/v1/llm-calls 端点, DevTools 代理 + UI 面板
- ModelSelector.envProvider 改为单例缓存, 避免重复创建 HTTP Client
- 新增 PurposeToolCalling 适配器, 后台思考工具调用走专用路由
- envFallback 超时 120s→180s, 显式设置 MaxRetries
- devtools.bat 全英文, 解决 Windows CMD GBK 编码乱码问题

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

backend/ai-core/internal/background/thinker.go

@@ -43,6 +43,7 @@ type Thinker struct {
 	personaLoader *persona.Loader
 	memRetriever  *memory.Retriever
 	llmAdapter    *llm.Adapter
+	toolAdapter   *llm.Adapter // 工具调用专用适配器
 	iotClient     *tools.IoTClient
 
 	// 记忆管理
@@ -224,6 +225,7 @@ func NewThinker(
 	personaLoader *persona.Loader,
 	memRetriever *memory.Retriever,
 	llmAdapter *llm.Adapter,
+	toolAdapter *llm.Adapter,
 	iotClient *tools.IoTClient,
 	memoryStore *memory.Store,
 	toolRegistry *tools.Registry,
@@ -237,6 +239,7 @@ func NewThinker(
 		personaLoader:       personaLoader,
 		memRetriever:        memRetriever,
 		llmAdapter:          llmAdapter,
+		toolAdapter:         toolAdapter,
 		iotClient:           iotClient,
 		thinkInterval:       cfg.ThinkInterval,
 		silenceTimeout:      cfg.SilenceTimeout,
@@ -598,7 +601,7 @@ func (t *Thinker) performThink(triggerReason string) {
 	var toolCallRecords []map[string]interface{}
 
 	for round := 0; round <= maxToolRounds; round++ {
-		resp, err := t.llmAdapter.ChatWithTools(ctx, messages, openAITools)
+		resp, err := t.toolAdapter.ChatWithTools(ctx, messages, openAITools)
 		if err != nil {
 			log.Printf("[后台思考] LLM调用失败 (round=%d): %v", round, err)
 			return