docs: fix README event count and Model javadoc to match code

README.md

@@ -55,7 +55,7 @@
 
 AgentScope Java 2.0 is a production-ready framework for building distributed, enterprise-grade agents, providing essential abstractions that work with rising model capability and built-in support for long-running, safely-controlled agent execution.
 
-- [**Event System** →](https://java.agentscope.io/v2/en/docs/building-blocks/message-and-event.html) A unified event stream with 28 typed events for real-time frontend rendering and human-in-the-loop.
+- [**Event System** →](https://java.agentscope.io/v2/en/docs/building-blocks/message-and-event.html) A unified event stream with 31 typed events for real-time frontend rendering and human-in-the-loop.
 - [**Permission System** →](https://java.agentscope.io/v2/en/docs/building-blocks/permission-system.html) Tool-call gating: allow / require user approval / deny.
 - [**Middleware** →](https://java.agentscope.io/v2/en/docs/building-blocks/middleware.html) AOP-style hook interception for flexibly extending the reasoning-acting loop.
 - [**Workspace & Sandbox** →](https://java.agentscope.io/v2/en/docs/harness/workspace.html) Run tools in isolated environments — local, Docker, Kubernetes, or AgentRun cloud sandbox.
@@ -190,7 +190,7 @@ Production agents must serve many tenants, run untrusted code safely, and surviv
 
 Messages, events, and the extension model are smaller, more orthogonal — HITL and event streaming are part of how the framework runs, not add-ons:
 
-- **Event stream** — 28 typed events covering model calls, text deltas, tool execution, and user confirmations in real time
+- **Event stream** — 31 typed events covering model calls, text deltas, tool execution, and user confirmations in real time
 - **Message model** — text / files / images / audio / video / tool results unified into `ContentBlock`, role-strict validation at construction
 - **Middleware** — `onAgent` / `onReasoning` / `onActing` / `onModelCall` / `onSystemPrompt` five stages replace v1's flat hooks
 - **HITL first class** — confirm tool arguments, approve sensitive actions, hand off to external systems, agent pauses and resumes exactly

README_zh.md

@@ -55,7 +55,7 @@
 
 AgentScope Java 2.0 是面向企业级、分布式、生产环境的智能体框架，提供与模型能力相匹配的核心 Harness 抽象，可支持长期、稳定、安全可控的智能体任务执行。
 
-- [**事件系统** →](https://java.agentscope.io/v2/zh/docs/building-blocks/message-and-event.html) 统一的事件流，28 种类型化事件，服务于前端实时渲染与 human-in-the-loop。
+- [**事件系统** →](https://java.agentscope.io/v2/zh/docs/building-blocks/message-and-event.html) 统一的事件流，31 种类型化事件，服务于前端实时渲染与 human-in-the-loop。
 - [**权限系统** →](https://java.agentscope.io/v2/zh/docs/building-blocks/permission-system.html) 工具调用决策机制：允许 / 用户审批 / 拒绝。
 - [**Middleware** →](https://java.agentscope.io/v2/zh/docs/building-blocks/middleware.html) 基于 AOP 模式的 hook 事件拦截机制，灵活扩展推理-行动循环。
 - [**Workspace 与沙箱** →](https://java.agentscope.io/v2/zh/docs/harness/workspace.html) 在隔离环境中执行工具 —— 本地、Docker、Kubernetes 或 AgentRun 云沙箱。

agentscope-core/src/main/java/io/agentscope/core/model/Model.java

@@ -73,7 +73,7 @@ default boolean supportsNativeStructuredOutputWithTools() {
     /**
      * Returns the model's context window size in tokens, or {@code 0} if unknown.
      *
-     * <p>Used by the compaction middleware to dynamically compute when to trigger
+     * <p>Used by the harness-layer compaction middleware to dynamically compute when to trigger
      * conversation summarization. Implementations should return the total context
      * window (input + output) for the configured model.
      *