fix(copilot): move get_user_tier import to top-level and expose cache via public API

- sdk/service.py: Move `get_user_tier` import from local (inside function)
  to module-level — no circular dependency exists.
- rate_limit.py: Expose `cache_clear`/`cache_delete` as attributes on the
  public `get_user_tier` function so callers never need to import the
  private `_fetch_user_tier`.
- rate_limit_test.py: Remove `_fetch_user_tier` import; use
  `get_user_tier.cache_clear()` instead.

autogpt_platform/backend/backend/copilot/rate_limit.py

@@ -407,10 +407,16 @@ async def get_user_tier(user_id: str) -> RateLimitTier:
     return DEFAULT_TIER
 
 
+# Expose cache management on the public function so callers (including tests)
+# never need to reach into the private ``_fetch_user_tier``.
+get_user_tier.cache_clear = _fetch_user_tier.cache_clear  # type: ignore[attr-defined]
+get_user_tier.cache_delete = _fetch_user_tier.cache_delete  # type: ignore[attr-defined]
+
+
 async def set_user_tier(user_id: str, tier: RateLimitTier) -> None:
     """Persist the user's rate-limit tier to the database.
 
-    Also invalidates the ``_fetch_user_tier`` cache for this user so that
+    Also invalidates the ``get_user_tier`` cache for this user so that
     subsequent rate-limit checks immediately see the new tier.
 
     Raises:
@@ -421,7 +427,7 @@ async def set_user_tier(user_id: str, tier: RateLimitTier) -> None:
         data={"rateLimitTier": tier.value},
     )
     # Invalidate cached tier so rate-limit checks pick up the change immediately.
-    _fetch_user_tier.cache_delete(user_id)
+    get_user_tier.cache_delete(user_id)  # type: ignore[attr-defined]
 
 
 async def get_global_rate_limits(

autogpt_platform/backend/backend/copilot/rate_limit_test.py

@@ -13,7 +13,6 @@ from .rate_limit import (
     RateLimitExceeded,
     RateLimitTier,
     UsageWindow,
-    _fetch_user_tier,
     check_rate_limit,
     get_global_rate_limits,
     get_usage_status,
@@ -388,8 +387,8 @@ class TestRateLimitTier:
 class TestGetUserTier:
     @pytest.fixture(autouse=True)
     def _clear_tier_cache(self):
-        """Clear the _fetch_user_tier cache before each test."""
-        _fetch_user_tier.cache_clear()
+        """Clear the get_user_tier cache before each test."""
+        get_user_tier.cache_clear()  # type: ignore[attr-defined]
 
     @pytest.mark.asyncio
     async def test_returns_tier_from_db(self):

autogpt_platform/backend/backend/copilot/sdk/service.py

@@ -33,6 +33,7 @@ from pydantic import BaseModel
 
 from backend.copilot.context import get_workspace_manager
 from backend.copilot.permissions import apply_tool_permissions
+from backend.copilot.rate_limit import get_user_tier
 from backend.data.redis_client import get_redis_async
 from backend.executor.cluster_lock import AsyncClusterLock
 from backend.util.exceptions import NotFoundError
@@ -1834,8 +1835,6 @@ async def stream_chat_completion_sdk(
         # langsmith tracing integration attaches them to every span.  This
         # is what Langfuse (or any OTEL backend) maps to its native
         # user/session fields.
-        from backend.copilot.rate_limit import get_user_tier
-
         _user_tier = await get_user_tier(user_id) if user_id else None
         _otel_metadata: dict[str, str] = {
             "resume": str(use_resume),