feat(memory): Day M.1 complete - importance scoring system

- Add FrequencyTracker: increment(), get_frequency_score(), get_recency_score(), get_time_decay()
- Add EmotionAnalyzer: EMOTION_KEYWORDS dict, extract(), calculate_score(), get_emotion_profile()
- Add ImpactEvaluator: evaluate(), get_topic_overlap(), rank_by_impact()
- Add ImportanceScorer: composite scoring (freq 35% + recency 20% + emotion 25% + impact 20%)
- Update UserMemory model: frequency_count, emotion_tags, importance_score, importance_level, associated_topics
- Integrate ImportanceScorer into memory_service.py (recall + importance update)
- Add 37 tests for all memory scoring components
- Fix urgency patterns: remove overly broad '今天' that matched neutral text
- Update memory-update checklist: mark all M.1 tasks complete

backend/app/models/memory.py

@@ -1,4 +1,15 @@
-from sqlalchemy import Column, String, Text, Integer, ForeignKey, Boolean, DateTime, Enum as SQLEnum
+from sqlalchemy import (
+    Column,
+    String,
+    Text,
+    Integer,
+    Float,
+    ForeignKey,
+    Boolean,
+    DateTime,
+    Enum as SQLEnum,
+    JSON,
+)
 from app.models.base import BaseModel, utc_now
 
 
@@ -7,12 +18,13 @@ class MemorySummary(BaseModel):
     对话摘要 — 中期记忆
     当一段对话超过阈值轮数时，自动生成摘要存入此表
     """
+
     __tablename__ = "memory_summaries"
 
     user_id = Column(String(36), ForeignKey("users.id"), nullable=False, index=True)
     conversation_id = Column(String(36), ForeignKey("conversations.id"), nullable=False, index=True)
-    summary_text = Column(Text, nullable=False)           # 摘要内容
-    turn_count = Column(Integer, default=0)                # 摘要时累计轮数
+    summary_text = Column(Text, nullable=False)  # 摘要内容
+    turn_count = Column(Integer, default=0)  # 摘要时累计轮数
     summary_at = Column(DateTime, default=utc_now, nullable=False)
 
 
@@ -21,14 +33,23 @@ class UserMemory(BaseModel):
     用户画像记忆 — 长期记忆
     从对话中提取的用户事实、偏好、目标
     """
+
     __tablename__ = "user_memories"
 
     user_id = Column(String(36), ForeignKey("users.id"), nullable=False, index=True)
-    memory_type = Column(String(50), nullable=False)       # fact | preference | goal | habit | other
-    content = Column(Text, nullable=False)                  # 记忆内容
-    importance = Column(Integer, default=5)                # 重要程度 1-10
-    is_recalled = Column(Boolean, default=False)           # 是否在当前对话中被召回
-    recall_count = Column(Integer, default=0)            # 被召回次数
+    memory_type = Column(String(50), nullable=False)  # fact | preference | goal | habit | other
+    content = Column(Text, nullable=False)  # 记忆内容
+    importance = Column(Integer, default=5)  # 重要程度 1-10 (legacy, replaced by importance_score)
+    is_recalled = Column(Boolean, default=False)  # 是否在当前对话中被召回
+    recall_count = Column(Integer, default=0)  # 被召回次数
     source_conversation_id = Column(String(36), nullable=True)  # 来源对话
     extracted_at = Column(DateTime, default=utc_now, nullable=False)
     last_recalled_at = Column(DateTime, nullable=True)
+    # M.1: 重要性评分系统
+    frequency_count = Column(
+        Integer, default=0
+    )  # 被召回次数 (duplicate of recall_count, for scoring clarity)
+    emotion_tags = Column(JSON, nullable=True)  # List of emotion keywords
+    importance_score = Column(Float, default=0.5)  # 重要性分数 0.0-1.0
+    importance_level = Column(String(20), default="medium")  # high | medium | low
+    associated_topics = Column(JSON, nullable=True)  # List of topic strings