feat: 上游 Key 唯一化、分组导入跳过、账号导入平台识别&远端校验&base_url 注入

- 上游 Key 命名改为 {prefix}-{upstream.id}-{safe_group_name}-{group_id}
- 唯一约束 (upstream_id, group_id, managed_prefix) 加 managed_prefix 列
- 上游检测成功时同步 Key 状态，远端已删/分组已删自动清理
- 重复分组导入跳过，目标网站已存在同名分组返回 exists
- 账号导入平台自动识别（auto/manual 模式）
- 全选可导入 Key 按钮 + 目标分组自动匹配
- 导入幂等：已导入过的 Key 校验远端账号，不存在则重建
- 新增同步接口 POST /sync-imported-upstream-keys
- account_exists() 通过拉取账号列表判断，避免 404 误判
- credentials.base_url 注入来源上游地址，避免 401
- 前端导入弹窗自动同步+刷新按钮+并发/优先级设置
- 新增 12 个测试覆盖同步、幂等、远端删除、校验失败路径

backend/app/database.py

@@ -26,10 +26,11 @@ def get_db():
 def init_db():
     """Create all tables."""
     # import models so SQLAlchemy registers them
-    from app.models import admin_user, upstream, snapshot, webhook_config, notification_log, custom_page, website, revoked_token  # noqa: F401
+    from app.models import admin_user, upstream, snapshot, webhook_config, notification_log, custom_page, website, revoked_token, upstream_key  # noqa: F401
     Base.metadata.create_all(bind=engine)
     _migrate_custom_pages()
     _migrate_upstreams()
+    _migrate_upstream_generated_keys()
 
 
 def _migrate_custom_pages():
@@ -87,3 +88,59 @@ def _migrate_upstreams():
         if "balance_divisor" not in columns:
             conn.execute(text("ALTER TABLE upstreams ADD COLUMN balance_divisor FLOAT NOT NULL DEFAULT 1.0"))
 
+
+def _migrate_upstream_generated_keys():
+    """Apply SQLite-safe migrations to the generated upstream keys table."""
+    inspector = inspect(engine)
+    if "upstream_generated_keys" not in inspector.get_table_names():
+        return
+    columns = {col["name"] for col in inspector.get_columns("upstream_generated_keys")}
+    with engine.begin() as conn:
+        if "imported_website_id" not in columns:
+            conn.execute(text("ALTER TABLE upstream_generated_keys ADD COLUMN imported_website_id INTEGER"))
+        if "imported_account_id" not in columns:
+            conn.execute(text("ALTER TABLE upstream_generated_keys ADD COLUMN imported_account_id VARCHAR(255)"))
+        if "imported_at" not in columns:
+            conn.execute(text("ALTER TABLE upstream_generated_keys ADD COLUMN imported_at DATETIME"))
+        if "updated_at" not in columns:
+            conn.execute(text("ALTER TABLE upstream_generated_keys ADD COLUMN updated_at DATETIME"))
+            conn.execute(text("UPDATE upstream_generated_keys SET updated_at = created_at WHERE updated_at IS NULL"))
+        if "managed_prefix" not in columns:
+            conn.execute(text("ALTER TABLE upstream_generated_keys ADD COLUMN managed_prefix VARCHAR(64)"))
+
+    # ——— 历史数据迁移：回填 managed_prefix + 清理重复 ———
+    with engine.begin() as conn:
+        # 1. 回填：key_name 以 SmartUp- 开头的旧记录设置 managed_prefix = 'SmartUp'
+        conn.execute(text(
+            "UPDATE upstream_generated_keys SET managed_prefix = 'SmartUp' "
+            "WHERE managed_prefix IS NULL AND key_name LIKE 'SmartUp-%'"
+        ))
+        # 2. 清理：同一 (upstream_id, group_id, managed_prefix) 只保留最新一条
+        #    SQLite 不支持子查询直接 DELETE，用两步
+        to_delete = conn.execute(text("""
+            SELECT id FROM upstream_generated_keys
+            WHERE managed_prefix IS NOT NULL
+            AND id NOT IN (
+                SELECT MAX(id) FROM upstream_generated_keys
+                WHERE managed_prefix IS NOT NULL
+                GROUP BY upstream_id, group_id, managed_prefix
+            )
+        """)).fetchall()
+        for (row_id,) in to_delete:
+            conn.execute(text("DELETE FROM upstream_generated_keys WHERE id = :id"), {"id": row_id})
+
+    # ——— 创建唯一索引 ———
+    try:
+        with engine.begin() as conn:
+            conn.execute(
+                text("CREATE UNIQUE INDEX IF NOT EXISTS uq_upstream_group_key "
+                     "ON upstream_generated_keys(upstream_id, group_id, key_name)")
+            )
+            conn.execute(
+                text("CREATE UNIQUE INDEX IF NOT EXISTS uq_upstream_group_managed "
+                     "ON upstream_generated_keys(upstream_id, group_id, managed_prefix) "
+                     "WHERE managed_prefix IS NOT NULL")
+            )
+    except Exception:
+        logger = __import__("logging").getLogger(__name__)
+        logger.warning("could not create unique indexes on upstream_generated_keys (non-fatal)")