refactor(proxy): 移除 unhealthy 状态、前端展示失败率替换连接数、命令提示修正

- 移除 ProxyNodeStatus.UNHEALTHY 枚举值，仅保留 online/offline
- 迁移脚本将已有 unhealthy 数据迁移为 offline，upgrade/downgrade 均有幂等性保护
- 前端代理节点列表用失败率列替换连接数列，超过 5% 高亮显示
- 节点列表排序从 updated_at DESC 改为 name ASC
- Rust 端命令行提示从 aether-proxy 改为 ./aether-proxy

src/services/proxy_node/service.py

@@ -392,13 +392,13 @@ class ProxyNodeService:
         query = db.query(ProxyNode)
         if status:
             normalized = status.strip().lower()
-            allowed = {"online", "unhealthy", "offline"}
+            allowed = {"online", "offline"}
             if normalized not in allowed:
                 raise InvalidRequestException(f"status 必须是以下之一: {sorted(allowed)}", "status")
             query = query.filter(ProxyNode.status == ProxyNodeStatus(normalized))
 
         total = query.count()
-        nodes = query.order_by(ProxyNode.updated_at.desc()).offset(skip).limit(limit).all()
+        nodes = query.order_by(ProxyNode.name.asc()).offset(skip).limit(limit).all()
         return nodes, total
 
     @staticmethod