remove legacy in-memory search state fallback

app/auth/rate_limit.py

@@ -1,6 +1,6 @@
 from __future__ import annotations
 
-from collections import defaultdict, deque
+from collections import deque
 from collections.abc import Callable
 from dataclasses import dataclass
 from math import ceil
@@ -25,14 +25,19 @@ class SlidingWindowRateLimiter:
         self._max_attempts = max_attempts
         self._window_seconds = window_seconds
         self._now = now
-        self._attempts: dict[str, deque[float]] = defaultdict(deque)
+        self._attempts: dict[str, deque[float]] = {}
         self._lock = Lock()
 
     def check(self, key: str) -> RateLimitDecision:
         with self._lock:
             now_value = self._now()
-            attempts = self._attempts[key]
+            attempts = self._attempts.get(key)
+            if attempts is None:
+                return RateLimitDecision(allowed=True)
             self._trim_expired(attempts, now_value)
+            if not attempts:
+                self._attempts.pop(key, None)
+                return RateLimitDecision(allowed=True)
             if len(attempts) >= self._max_attempts:
                 retry_after = self._window_seconds - (now_value - attempts[0])
                 return RateLimitDecision(
@@ -44,7 +49,10 @@ class SlidingWindowRateLimiter:
     def record_failure(self, key: str) -> None:
         with self._lock:
             now_value = self._now()
-            attempts = self._attempts[key]
+            attempts = self._attempts.get(key)
+            if attempts is None:
+                attempts = deque()
+                self._attempts[key] = attempts
             self._trim_expired(attempts, now_value)
             attempts.append(now_value)
 
@@ -59,4 +67,3 @@ class SlidingWindowRateLimiter:
     def _trim_expired(self, attempts: deque[float], now_value: float) -> None:
         while attempts and now_value - attempts[0] >= self._window_seconds:
             attempts.popleft()
-