feat(llm): 新增 IR 类型层 + 切换 LlmProvider trait 签名

引入跨 Provider 统一的新类型系统：
- Message 扁平大枚举（System/User/UserImage/Assistant/ToolResult）
- MessageRequest + MessageResponse + StreamEvent（高精度 IR）
- PartialMessageResponse 含 apply_to/finalize 汇聚算法
- ProviderType enum 扩展 + ProviderCapabilities

切换 LlmProvider trait 签名为 chat(MessageRequest) → MessageResponse，
chat_stream 返回 Stream<Item = Result<StreamEvent, LlmError>>，新增 capabilities()。

StreamEvent 命名冲突处理：旧变体迁移至 old_stream::LegacyStreamEvent，
stream.rs 通过 pub use 重新导出新高精度事件。

OpenaiProvider 添加 Phase 0 临时桥接（MessageRequest ↔ OpenaiChatRequest 转换），
标注 ponytail: Phase 0 临时桥接 标记，Phase 1 重写时移除。

测试 147 passed / 0 failed（新增 31 个新类型测试）。

src/agent/session.rs

@@ -15,7 +15,9 @@ use crate::agent::runtime::RuntimeBundle;
 use crate::agent::session_memory::SessionMemory;
 use crate::llm::cycle::{CostTracker, CycleConfig, LlmCycle};
 use crate::llm::hooks::{HookContext, HookEvent};
-use crate::llm::types::ChatResponse;
+use crate::llm::types::message::{ContentBlock, Message};
+use crate::llm::types::openai_message::OpenaiChatMessage;
+use crate::llm::types::response_v2::MessageResponse;
 use crate::memory::store::InMemoryStore;
 
 /// Agent 会话实例。
@@ -120,7 +122,7 @@ impl AgentSession {
     pub async fn submit_turn(
         &mut self,
         user_input: impl Into<String>,
-    ) -> Result<ChatResponse, AgentError> {
+    ) -> Result<MessageResponse, AgentError> {
         let turn_index = self.turn_index;
         let hook_executor = Arc::clone(&self.bundle.hook_executor);
 
@@ -168,13 +170,15 @@ impl AgentSession {
 mod tests {
     use super::*;
     use crate::agent::builder::AgentBuilder;
+    use crate::llm::error::LlmError;
     use crate::llm::hooks::{Hook, HookContext, HookExecutor, HookResult};
-    use crate::llm::provider::LlmProvider;
-    use crate::llm::types::{
-        ChatRequest, ChatResponse, FinishReason, OpenaiChatMessage,
-    };
+    use crate::llm::provider::{LlmProvider, ProviderCapabilities, ProviderFeatures};
+    use crate::llm::types::request_v2::MessageRequest;
+    use crate::llm::types::response_v2::{MessageResponse, StopReason, StreamEvent};
     use crate::tools::ToolRegistry;
     use async_trait::async_trait;
+    use futures_core::Stream;
+    use std::pin::Pin;
     use std::sync::atomic::{AtomicU32, Ordering};
 
     /// 计数 hook —— 每被调用一次 +1。
@@ -200,11 +204,11 @@ mod tests {
 
     /// MockProvider：按调用顺序返回预设响应。
     struct MockProvider {
-        responses: std::sync::Mutex<Vec<ChatResponse>>,
+        responses: std::sync::Mutex<Vec<MessageResponse>>,
     }
 
     impl MockProvider {
-        fn new(responses: Vec<ChatResponse>) -> Self {
+        fn new(responses: Vec<MessageResponse>) -> Self {
             Self {
                 responses: std::sync::Mutex::new(responses),
             }
@@ -213,15 +217,27 @@ mod tests {
 
     #[async_trait]
     impl LlmProvider for MockProvider {
-        async fn chat(&self, _request: ChatRequest) -> Result<ChatResponse, crate::llm::error::LlmError> {
+        async fn chat(&self, _request: MessageRequest) -> Result<MessageResponse, LlmError> {
             let mut responses = self.responses.lock().unwrap();
             if responses.is_empty() {
-                return Err(crate::llm::error::LlmError::Other(
-                    "no more mock responses".into(),
-                ));
+                return Err(LlmError::Other("no more mock responses".into()));
             }
             Ok(responses.remove(0))
         }
+        async fn chat_stream(
+            &self,
+            _request: MessageRequest,
+        ) -> Result<Pin<Box<dyn Stream<Item = Result<StreamEvent, LlmError>> + Send>>, LlmError>
+        {
+            unimplemented!()
+        }
+        fn capabilities(&self) -> ProviderCapabilities {
+            ProviderCapabilities {
+                provider_name: "mock",
+                supported_models: None,
+                features: ProviderFeatures::default(),
+            }
+        }
     }
 
     struct StubAgent {
@@ -238,11 +254,18 @@ mod tests {
         }
     }
 
-    fn assistant_text(text: &str) -> ChatResponse {
-        ChatResponse {
-            message: OpenaiChatMessage::assistant_text(text),
+    fn assistant_text(text: &str) -> MessageResponse {
+        MessageResponse {
+            id: String::new(),
+            model: String::new(),
+            message: Message::Assistant {
+                content: vec![ContentBlock::Text {
+                    text: text.into(),
+                }],
+            },
             usage: crate::llm::types::Usage::from_input_output(10, 5),
-            stop_reason: Some(FinishReason::Stop),
+            stop_reason: StopReason::Stop,
+            extra: std::collections::HashMap::new(),
         }
     }
 
@@ -267,17 +290,7 @@ mod tests {
         assert_eq!(session.turn_index(), 0);
 
         let response = session.submit_turn("hi").await.unwrap();
-        let text = match &response.message {
-            OpenaiChatMessage::Assistant { content, .. } => {
-                if let crate::llm::types::ContentField::String(s) = content {
-                    s.clone()
-                } else {
-                    String::new()
-                }
-            }
-            _ => String::new(),
-        };
-        assert_eq!(text, "hello back");
+        assert_eq!(response.text(), "hello back");
         assert_eq!(session.turn_index(), 1);
         assert_eq!(session.usage().total().prompt_tokens, 10);
         assert_eq!(session.usage().total().completion_tokens, 5);