feat(pool): 引入多维评分调度策略与账号状态检测

- 新增 multi_score 调度模式，支持 LRU/延迟/健康度/剩余额度多维加权评分
- 新增调度预设维度系统（free_team_first, quota_balanced, recent_refresh, single_account），支持有序对象列表配置格式并兼容旧字符串列表
- 新增 account_state 模块，统一账号封禁/受限检测逻辑，替代分散在 routes 中的判断代码
- 新增 health_cache 模块和 latency 采样（redis_ops.record_latency / batch_get_latency_avgs）
- RequestDispatcher 返回 ttfb_ms，PoolManager.on_request_success 记录延迟样本
- 前端：PoolConfigDialog 替换为 PoolSchedulingDialog，支持预设维度可视化配置；号池管理页增加调度模式标签与账号异常 Badge 显示
- 提取前端 accountBlock 工具函数，ProviderDetailDrawer 复用统一判断
- scheduling_dimensions 增加 account_state 和 latency 维度评估
- 补充 account_state、health_cache、multi_score 策略、preset 维度、redis latency 等测试

tests/services/test_pool_preset_dimensions.py

@@ -0,0 +1,83 @@
+"""Tests for pool preset dimension registry and built-in dimensions."""
+
+from __future__ import annotations
+
+from types import SimpleNamespace
+
+import src.services.provider.pool.dimensions  # noqa: F401
+from src.services.provider.pool.dimensions import get_preset_dimension, get_preset_names
+
+
+def _key(metadata: dict, *, plan_type: str | None = None) -> SimpleNamespace:
+    return SimpleNamespace(upstream_metadata=metadata, oauth_plan_type=plan_type)
+
+
+def test_registry_discovers_builtin_dimensions() -> None:
+    names = get_preset_names()
+    assert {"free_team_first", "recent_refresh", "quota_balanced", "single_account"}.issubset(names)
+
+
+def test_universal_dimensions_are_applicable_to_any_provider() -> None:
+    for name in ("quota_balanced", "single_account"):
+        dim = get_preset_dimension(name)
+        assert dim is not None
+        assert dim.is_applicable("openai") is True
+        assert dim.is_applicable("codex") is True
+        assert dim.is_applicable("unknown_provider") is True
+
+
+def test_provider_specific_dimensions_are_filtered() -> None:
+    for name in ("free_team_first", "recent_refresh"):
+        dim = get_preset_dimension(name)
+        assert dim is not None
+        assert dim.is_applicable("codex") is True
+        assert dim.is_applicable("kiro") is True
+        assert dim.is_applicable("openai") is False
+
+
+def test_builtin_dimensions_compute_metric_in_range() -> None:
+    all_key_ids = ["k1", "k2", "k3"]
+    lru_scores = {"k1": 100.0, "k2": 800.0, "k3": 300.0}
+    keys_by_id = {
+        "k1": _key(
+            {
+                "codex": {
+                    "plan_type": "plus",
+                    "primary_reset_seconds": 900,
+                    "primary_used_percent": 60,
+                }
+            }
+        ),
+        "k2": _key(
+            {
+                "codex": {
+                    "plan_type": "free",
+                    "primary_reset_seconds": 120,
+                    "primary_used_percent": 20,
+                }
+            }
+        ),
+        "k3": _key(
+            {
+                "kiro": {
+                    "next_reset_at": 4102444800,
+                    "usage_percentage": 45,
+                    "subscription_title": "Kiro Team",
+                }
+            },
+            plan_type="team",
+        ),
+    }
+
+    for name in get_preset_names():
+        dim = get_preset_dimension(name)
+        assert dim is not None
+        mode = dim.default_mode
+        metric = dim.compute_metric(
+            key_id="k1",
+            all_key_ids=all_key_ids,
+            keys_by_id=keys_by_id,
+            lru_scores=lru_scores,
+            mode=mode,
+        )
+        assert 0.0 <= metric <= 1.0