feat: AI市场分析功能

- 后端：AI分析服务、工具调用、定时任务
- 前端：报告列表、详情页、Mermaid图表支持
- 支持OpenAI API兼容模型
- 收盘后自动分析生成报告

backend/routes/ai_analysis.py

@@ -0,0 +1,127 @@
+"""AI 分析报告路由（/api/ai-analysis）
+
+提供：
+- 报告列表
+- 报告详情
+- 手动触发分析（12小时频率限制）
+- 重新生成报告
+- 检查某日是否有报告
+"""
+
+import json
+from datetime import datetime, timezone, timedelta
+
+from fastapi import APIRouter, HTTPException
+from fastapi.responses import JSONResponse
+
+from database import get_connection, dict_from_row
+
+_CST = timezone(timedelta(hours=8))
+_NO_CACHE_HEADERS = {"Cache-Control": "no-store, no-cache, must-revalidate, max-age=0"}
+
+router = APIRouter()
+
+
+def _has_ai_report(trade_date: str) -> bool:
+    conn = get_connection()
+    try:
+        row = conn.execute(
+            "SELECT 1 FROM ai_reports WHERE trade_date = ? AND report_type = 'daily' LIMIT 1",
+            (trade_date,)
+        ).fetchone()
+        return row is not None
+    finally:
+        conn.close()
+
+
+def _can_trigger(trade_date: str) -> tuple[bool, str]:
+    """检查是否可以触发分析（12小时频率限制）"""
+    conn = get_connection()
+    try:
+        row = conn.execute(
+            "SELECT created_at FROM ai_reports WHERE trade_date = ? ORDER BY id DESC LIMIT 1",
+            (trade_date,)
+        ).fetchone()
+        if row:
+            last_time = datetime.strptime(row["created_at"], "%Y-%m-%d %H:%M:%S")
+            now = datetime.now(_CST)
+            if (now - last_time).total_seconds() < 12 * 3600:
+                return False, "12小时内已触发过，请稍后再试"
+        return True, ""
+    finally:
+        conn.close()
+
+
+@router.get("/ai-analysis", summary="AI 分析报告列表")
+async def list_reports():
+    conn = get_connection()
+    try:
+        rows = conn.execute(
+            "SELECT id, trade_date, report_type, title, summary, tools_used, model, tokens_used, created_at "
+            "FROM ai_reports ORDER BY trade_date DESC, id DESC LIMIT 50"
+        ).fetchall()
+        items = []
+        for r in rows:
+            item = dict_from_row(r)
+            item["toolsUsed"] = json.loads(item.pop("tools_used") or "[]")
+            items.append(item)
+        return JSONResponse({"data": items}, headers=_NO_CACHE_HEADERS)
+    finally:
+        conn.close()
+
+
+@router.get("/ai-analysis/{report_id}", summary="AI 分析报告详情")
+async def get_report(report_id: int):
+    conn = get_connection()
+    try:
+        row = conn.execute("SELECT * FROM ai_reports WHERE id = ?", (report_id,)).fetchone()
+        if not row:
+            raise HTTPException(status_code=404, detail="报告不存在")
+        item = dict_from_row(row)
+        item["toolsUsed"] = json.loads(item.pop("tools_used") or "[]")
+        return JSONResponse({"data": item}, headers=_NO_CACHE_HEADERS)
+    finally:
+        conn.close()
+
+
+@router.get("/ai-analysis/check/{trade_date}", summary="检查某日是否有 AI 分析报告")
+async def check_report(trade_date: str):
+    has = _has_ai_report(trade_date)
+    return JSONResponse({"data": {"hasReport": has, "tradeDate": trade_date}})
+
+
+@router.post("/ai-analysis/trigger", summary="手动触发今日 AI 分析")
+async def trigger_analysis():
+    now = datetime.now(_CST)
+    trade_date = now.strftime("%Y-%m-%d")
+
+    if _has_ai_report(trade_date):
+        raise HTTPException(status_code=400, detail="今日已有分析报告")
+
+    can, msg = _can_trigger(trade_date)
+    if not can:
+        raise HTTPException(status_code=429, detail=msg)
+
+    from services.ai_service import collect_ai_analysis
+    result = await collect_ai_analysis(trade_date)
+    return JSONResponse({"data": result})
+
+
+@router.post("/ai-analysis/{report_id}/regenerate", summary="重新生成 AI 分析报告")
+async def regenerate_report(report_id: int):
+    conn = get_connection()
+    try:
+        row = conn.execute("SELECT trade_date FROM ai_reports WHERE id = ?", (report_id,)).fetchone()
+        if not row:
+            raise HTTPException(status_code=404, detail="报告不存在")
+        trade_date = row["trade_date"]
+    finally:
+        conn.close()
+
+    can, msg = _can_trigger(trade_date)
+    if not can:
+        raise HTTPException(status_code=429, detail=msg)
+
+    from services.ai_service import collect_ai_analysis
+    result = await collect_ai_analysis(trade_date)
+    return JSONResponse({"data": result})