refactor: 统一自建 Tool Search — 移除 tool_reference/defer_loading 依赖，全 provider 通用

- 重命名 ExecuteTool → ExecuteExtraTool，作为一等工具始终可用
- ToolSearchTool 输出改为纯文本（区分 core/deferred），移除 tool_reference blocks
- 移除 modelSupportsToolReference() 及相关 GrowthBook 配置
- 移除 API 侧 defer_loading 字段和 tool search beta header 注入
- 简化 system prompt（工具使用指南从 ~120 行压缩到 ~10 行）
- extractDiscoveredToolNames 支持文本格式解析（向后兼容旧 session 的 tool_reference）
- 更新 promptEngineeringAudit 测试以匹配简化后的 prompt 结构

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

src/utils/api.ts

@@ -230,11 +230,7 @@ export async function toolToAPISchema(
   }
 
   // CLAUDE_CODE_DISABLE_EXPERIMENTAL_BETAS is the kill switch for beta API
-  // shapes. Proxy gateways (ANTHROPIC_BASE_URL → LiteLLM → Bedrock) reject
-  // fields like defer_loading with "Extra inputs are not permitted". The gates
-  // above each field are scattered and not all provider-aware, so this strips
-  // everything not in the base-tool allowlist at the one choke point all tool
-  // schemas pass through — including fields added in the future.
+  // shapes. Strips defer_loading and other beta fields from tool schemas.
   // cache_control is allowlisted: the base {type: 'ephemeral'} shape is
   // standard prompt caching (Bedrock/Vertex supported); the beta sub-fields
   // (scope, ttl) are already gated upstream by shouldIncludeFirstPartyOnlyBetas