fix(graph_sqlite): fix ExpandFromResults - wrong field access + Chinese entity extraction

Root causes fixed:
1. ExpandFromResults accessed p["to"] but SQLiteGraphStore.Navigate
   returns {"source","target","relation","weight"} - field name mismatch
   caused graph expansion to always return empty
2. Original logic checked seen[r.ID] (memory ID) against entity name
   (neighbor string), which never matched - removed stale logic

Also improved extractPotentialEntities:
- Now handles Chinese continuous character sequences (2-8 chars)
- Chinese entities extracted as whole words, not split by spaces
- English capitalized words preserved with original casing

Changes:
- ExpandFromResults: correct field access (source/target/relation),
  bidirectional neighbor detection (forward + reverse edges),
  separate seenEntities map for entity dedup
- extractPotentialEntities: full rewrite for Chinese + better English
  word tokenization without space-based splitting

go/internal/governance/graph_sqlite.go

@@ -556,10 +556,11 @@ func (gs *SQLiteGraphStore) GetGraph(namespace string, limit int) ([]map[string]
 
 func (gs *SQLiteGraphStore) ExpandFromResults(results []models.RecallResult, namespace string, maxHops int) []models.RecallResult {
 	// 从 recall 结果提取实体，展开图谱邻居
-	expanded := make([]models.RecallResult, len(results))
-	copy(expanded, results)
+	expanded := make([]models.RecallResult, 0, len(results)*2)
+	expanded = append(expanded, results...)
 
-	seen := make(map[string]bool)
+	seen := make(map[string]bool) // 追踪已访问的记忆 ID
+	seenEntities := make(map[string]bool) // 追踪已访问的实体（用于去重扩展结果）
 	for _, r := range results {
 		if seen[r.ID] {
 			continue
@@ -569,53 +570,93 @@ func (gs *SQLiteGraphStore) ExpandFromResults(results []models.RecallResult, nam
 		// 从内容中提取可能作为实体的关键词
 		entities := extractPotentialEntities(r.Content)
 		for _, entity := range entities {
+			if seenEntities[entity] {
+				continue
+			}
+			seenEntities[entity] = true
+
 			nodeID := normalizeEntityID(entity)
 			paths, _ := gs.Navigate(nodeID, maxHops, namespace)
 			for _, p := range paths {
-				if target, ok := p["to"].(string); ok && !seen[target] {
-					seen[target] = true
-					expanded = append(expanded, models.RecallResult{
-						Content: fmt.Sprintf("[graph] %s --[%s]--> %s", entity, p["relation"], p["to"]),
-						Score:   r.Score * 0.5,
-					})
+				// Navigate 返回的是单条边 (source/target/relation/weight)
+				// 有两种情况：
+				//   1. source == entity（正向边）：target 是下游邻居
+				//   2. target == entity（反向边）：source 是上游邻居
+				var neighbor, rel string
+				src, _ := p["source"].(string)
+				tgt, _ := p["target"].(string)
+				relVal, _ := p["relation"].(string)
+				if src == entity && tgt != "" {
+					neighbor = tgt
+					rel = relVal
+				} else if tgt == entity && src != "" {
+					neighbor = src
+					rel = "↩ " + relVal
 				}
+				if neighbor == "" {
+					continue
+				}
+				if seenEntities[neighbor] {
+					continue
+				}
+				seenEntities[neighbor] = true
+				expanded = append(expanded, models.RecallResult{
+					Content: fmt.Sprintf("[graph] %s --[%s]--> %s", entity, rel, neighbor),
+					Score:   r.Score * 0.5,
+				})
 			}
 		}
 	}
 	return expanded
 }
 
-// extractPotentialEntities 从文本中提取可能作为图谱实体的关键词
+// extractPotentialEntities 从文本中提取可能作为图谱实体的关键词（支持中文连续字符）
 func extractPotentialEntities(text string) []string {
 	var entities []string
 	seen := make(map[string]bool)
-	for _, w := range strings.Fields(text) {
+	runes := []rune(text)
+	for i := 0; i < len(runes); {
+		r := runes[i]
+		// 中文字符：收集连续的中文字符序列（2-8字）
+		if r >= 0x4E00 && r <= 0x9FFF {
+			start := i
+			i++
+			for i < len(runes) && runes[i] >= 0x4E00 && runes[i] <= 0x9FFF {
+				i++
+			}
+			chinese := string(runes[start:i])
+			// 不等式：2 <= len(chinese) <= 8
+			if len(chinese) >= 2 && len(chinese) <= 8 && !seen[chinese] {
+				seen[chinese] = true
+				entities = append(entities, chinese)
+			}
+			continue
+		}
+		// 非中文字符，收集整个单词
+		start := i
+		for i < len(runes) {
+			r2 := runes[i]
+			if r2 >= 0x4E00 && r2 <= 0x9FFF {
+				break // 遇到汉字则停止
+			}
+			i++
+		}
+		if i-start < 2 {
+			continue
+		}
+		w := string(runes[start:i])
+		// 去除首尾标点
 		w = strings.Trim(w, ",.;:!?，。；：！？、\"'()（）[]【】")
 		if len(w) < 2 {
 			continue
 		}
-		// 大写开头（英文命名实体）
-		runes := []rune(w)
-		if len(runes) >= 2 && runes[0] >= 'A' && runes[0] <= 'Z' {
-			if !seen[strings.ToLower(w)] {
-				seen[strings.ToLower(w)] = true
-				entities = append(entities, w)
-			}
-		}
-		// 纯中文字符 2-8 字
-		chineseOnly := true
-		chineseRunes := 0
-		for _, r := range runes {
-			if r >= 0x4E00 && r <= 0x9FFF {
-				chineseRunes++
-			} else {
-				chineseOnly = false
-			}
-		}
-		if chineseOnly && chineseRunes >= 2 && chineseRunes <= 8 {
-			if !seen[w] {
-				seen[w] = true
-				entities = append(entities, w)
+		// 英文大写字母开头的词
+		first := []rune(w)
+		if len(first) > 0 && first[0] >= 'A' && first[0] <= 'Z' {
+			lower := strings.ToLower(w)
+			if !seen[lower] {
+				seen[lower] = true
+				entities = append(entities, w) // 保留原始大小写
 			}
 		}
 	}