refactor: 补全services层+消除core/models.py冗余+修复network.py缩进错误

高优先级问题1 - Router层业务逻辑泄漏:
- 新建 services/proxy_service.py: 从routers/proxy.py提取ProxyService类，
  封装代理/白名单测试逻辑、全局状态管理、所有core/ import
- 新建 services/account_service.py: 从routers/accounts.py提取
  split_account_line、cookie_account_ids_query、parse_and_build_accounts，
  封装core.douyu.email_verifier的import
- 重写routers/proxy.py为薄HTTP/WS层，移除所有core/直接依赖
- 精简routers/accounts.py，从account_service导入业务函数
- 将_active_batches全局状态移入login_service.py的BatchRegistry类
- routers/login.py改为使用batch_registry单例

高优先级问题2 - 两套models.py冗余:
- 删除core/models.py（Account dataclass + ProxyConfig死代码）
- 在core/douyu/login.py新增AccountLike Protocol替代Account dataclass
- login_service.py中Account(...)构造改为SimpleNamespace(...)
- 清理core/__init__.py导出

附带修复:
- core/geetest/common/network.py的get_picture()函数缩进错误
  （缺少if match:块和data赋值语句）

web/backend/services/account_service.py

@@ -0,0 +1,76 @@
+"""账号管理服务层：封装账号导入、查询等业务逻辑，隔离 core/ 依赖。"""
+
+import csv
+import re
+
+from sqlalchemy import func
+from sqlalchemy.orm import Session, joinedload
+
+from ..models import Account, AuditLog, LoginTask
+from ..permissions import user_has_permission
+
+
+EMAIL_PATTERN = re.compile(r'^[^\s@|]+@[^\s@|]+\.[^\s@|]+$')
+
+
+def split_account_line(line: str) -> list[str]:
+    """拆分一行账号文本，支持 |、tab、逗号、空格分隔。"""
+    if '|' in line:
+        return line.split('|')
+    if '\t' in line:
+        return line.split('\t')
+    if ',' in line:
+        return next(csv.reader([line]))
+    return line.split()
+
+
+def cookie_account_ids_query(db: Session):
+    """返回有成功登录记录(cookie非空)的账号ID子查询。"""
+    return db.query(LoginTask.account_id).filter(
+        LoginTask.status == 'success',
+        LoginTask.cookie != '',
+        LoginTask.cookie.isnot(None),
+    ).distinct()
+
+
+def parse_and_build_accounts(text: str) -> tuple[list[Account], int]:
+    """
+    解析批量导入文本，构建 Account ORM 对象列表。
+
+    Returns:
+        (accounts, skipped_count)
+    """
+    from core.douyu.email_verifier import get_email_config_for_account
+
+    accounts = []
+    skipped = 0
+    for line in text.strip().split('\n'):
+        line = line.strip()
+        if not line or line.startswith('#'):
+            continue
+        parts = split_account_line(line)
+        if len(parts) < 4:
+            skipped += 1
+            continue
+
+        username, password, email, email_password = [p.strip() for p in parts[:4]]
+        tag = parts[4].strip() if len(parts) > 4 else ""
+        if not all([username, password, email, email_password]):
+            skipped += 1
+            continue
+        if not EMAIL_PATTERN.match(email):
+            skipped += 1
+            continue
+
+        email_cfg = get_email_config_for_account(email)
+        accounts.append(Account(
+            username=username,
+            password=password,
+            email=email,
+            email_password=email_password,
+            email_imap_server=email_cfg['server'],
+            email_imap_port=email_cfg.get('port', 993),
+            email_imap_ssl=email_cfg.get('ssl', True),
+            tag=tag,
+        ))
+    return accounts, skipped