fix(ai): 修复自动分析管道 — 补全建议生成 + 事件发布

自动分析批处理（auto_analysis.rs）在完成流式分析后仅保存结果，
缺少三个关键步骤导致关怀引擎无法启动：
1. 不解析双通道输出（StructuredOutput）→ 无结构化建议
2. 不调用 SuggestionService.create_suggestions() → 无建议记录
3. 不发布 ai.analysis.completed 事件 → 下游消费者无感知

修复方案：提取 post_process_analysis() 共享函数，统一处理
解析→创建建议→发布事件的后处理逻辑，SSE handler 和自动分析共用。

crates/erp-ai/src/handler/mod.rs

@@ -9,7 +9,6 @@ use serde::Deserialize;
 use std::convert::Infallible;
 
 use crate::dto::{AnalysisSseEvent, AnalysisType};
-use crate::service::suggestion::SuggestionService;
 use crate::state::AiState;
 
 pub mod suggestion_handler;
@@ -534,49 +533,19 @@ fn build_sse_stream(
         }
 
         let metadata = serde_json::json!({"analysis_type": analysis_type});
-        let _ = state.analysis.complete_analysis(analysis_id, full_content.clone(), metadata).await;
+        let _ = state.analysis.complete_analysis(analysis_id, full_content.clone(), metadata.clone()).await;
 
-        // 解析双通道输出并创建建议记录
+        // 后处理：解析双通道输出、创建建议、发布事件
-        let parsed = crate::service::output_parser::parse_dual_channel(&full_content).unwrap_or(
+        crate::service::post_process::post_process_analysis(
-            crate::dto::suggestion::ParsedOutput {
+            &state,
-                text_content: full_content.clone(),
+            analysis_id,
-                structured: None,
+            &full_content,
-            },
-        );
-
-        let mut event_payload = serde_json::json!({
-            "analysis_id": analysis_id,
-            "analysis_type": analysis_type,
-            "patient_id": patient_id,
-            "doctor_id": doctor_id,
-        });
-
-        if let Some(ref structured) = parsed.structured {
-            event_payload["risk_level"] = serde_json::json!(structured.risk_level.as_str());
-            event_payload["suggestion_count"] = serde_json::json!(structured.suggestions.len());
-
-            if !structured.suggestions.is_empty() {
-                let _ = SuggestionService::create_suggestions(
-                    &state.db,
-                    tenant_id,
-                    analysis_id,
-                    &structured.suggestions,
-                    structured.risk_level,
-                    &structured.baseline_summary,
-                    Some(doctor_id),
-                ).await;
-            }
-        } else {
-            let _ = SuggestionService::mark_parse_failed(&state.db, analysis_id).await;
-        }
-
-        // 发布 AI 分析完成事件
-        let event = erp_core::events::DomainEvent::new(
-            "ai.analysis.completed",
             tenant_id,
-            erp_core::events::build_event_payload(event_payload),
+            patient_id,
-        );
+            doctor_id,
-        state.event_bus.publish(event, &state.db).await;
+            analysis_type,
+            metadata,
+        ).await;
 
         let done_event = AnalysisSseEvent::Done {
             analysis_id,

crates/erp-ai/src/service/auto_analysis.rs

@@ -208,11 +208,33 @@ async fn analyze_tenant_high_risk_patients(
                     });
                     if let Err(e) = state
                         .analysis
-                        .complete_analysis(analysis_id, full_content, metadata)
+                        .complete_analysis(analysis_id, full_content.clone(), metadata.clone())
                         .await
                     {
                         tracing::warn!(error = %e, "保存分析结果失败");
                     }
+
+                    // 后处理：解析双通道输出、创建建议、发布事件
+                    let result = super::post_process::post_process_analysis(
+                        state,
+                        analysis_id,
+                        &full_content,
+                        tenant_id,
+                        patient_id,
+                        system_user_id,
+                        "trend",
+                        metadata,
+                    )
+                    .await;
+
+                    tracing::info!(
+                        patient_id = %patient_id,
+                        analysis_id = %analysis_id,
+                        risk_level = ?result.risk_level,
+                        suggestion_count = result.suggestion_count,
+                        "自动分析后处理完成"
+                    );
+
                     analyzed += 1;
                 }
             }

crates/erp-ai/src/service/mod.rs

@@ -3,6 +3,7 @@ pub mod auto_analysis;
 pub mod comparison;
 pub mod local_rules;
 pub mod output_parser;
+pub mod post_process;
 pub mod prompt;
 pub mod reanalysis;
 pub mod suggestion;

crates/erp-ai/src/service/post_process.rs

@@ -0,0 +1,92 @@
+//! 分析后处理 — 解析双通道输出、创建建议、发布事件
+//!
+//! 被 SSE handler（`build_sse_stream`）和自动分析（`auto_analysis`）共用。
+
+use uuid::Uuid;
+
+use crate::dto::suggestion::ParsedOutput;
+use crate::service::output_parser;
+use crate::service::suggestion::SuggestionService;
+use crate::state::AiState;
+
+/// 分析后处理结果
+pub struct PostProcessResult {
+    pub risk_level: Option<String>,
+    pub suggestion_count: usize,
+    pub suggestion_ids: Vec<Uuid>,
+}
+
+/// 对完成的分析执行后处理：解析双通道输出、创建建议、发布事件
+pub async fn post_process_analysis(
+    state: &AiState,
+    analysis_id: Uuid,
+    full_content: &str,
+    tenant_id: Uuid,
+    patient_id: Uuid,
+    user_id: Uuid,
+    analysis_type: &str,
+    _metadata: serde_json::Value,
+) -> PostProcessResult {
+    // 1. 解析双通道输出
+    let parsed = output_parser::parse_dual_channel(full_content).unwrap_or(ParsedOutput {
+        text_content: full_content.to_string(),
+        structured: None,
+    });
+
+    // 2. 构建事件 payload
+    let mut event_payload = serde_json::json!({
+        "analysis_id": analysis_id,
+        "analysis_type": analysis_type,
+        "patient_id": patient_id,
+        "doctor_id": user_id,
+    });
+
+    let mut risk_level_str: Option<String> = None;
+    let mut suggestion_ids = Vec::new();
+
+    if let Some(ref structured) = parsed.structured {
+        risk_level_str = Some(structured.risk_level.as_str().to_string());
+        event_payload["risk_level"] = serde_json::json!(structured.risk_level.as_str());
+        event_payload["suggestion_count"] = serde_json::json!(structured.suggestions.len());
+
+        // 3. 创建建议记录
+        if !structured.suggestions.is_empty() {
+            match SuggestionService::create_suggestions(
+                &state.db,
+                tenant_id,
+                analysis_id,
+                &structured.suggestions,
+                structured.risk_level,
+                &structured.baseline_summary,
+                Some(user_id),
+            )
+            .await
+            {
+                Ok(ids) => suggestion_ids = ids,
+                Err(e) => {
+                    tracing::warn!(
+                        analysis_id = %analysis_id,
+                        error = %e,
+                        "自动分析创建建议记录失败"
+                    );
+                }
+            }
+        }
+    } else {
+        let _ = SuggestionService::mark_parse_failed(&state.db, analysis_id).await;
+    }
+
+    // 4. 发布 AI 分析完成事件
+    let event = erp_core::events::DomainEvent::new(
+        "ai.analysis.completed",
+        tenant_id,
+        erp_core::events::build_event_payload(event_payload),
+    );
+    state.event_bus.publish(event, &state.db).await;
+
+    PostProcessResult {
+        risk_level: risk_level_str,
+        suggestion_count: suggestion_ids.len(),
+        suggestion_ids,
+    }
+}