test: cover config-level disable for xhigh/max and the effort-only load warning

- ApplyReasoningEffort keeps an operator's reasoning.disable for high, xhigh and max
- sglang: _configured_budget() feeds the strict-thinking warning; tested with
  effort_budgets alone

Signed-off-by: pos-ei-don <1822533+pos-ei-don@users.noreply.github.com>
(cherry picked from commit 45a960ca1f)

backend/python/sglang/backend.py

@@ -207,6 +207,10 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             return self.effort_budgets[effort]
         return self.thinking_budget
 
+    def _configured_budget(self) -> Optional[int]:
+        """Any budget configured for this model (fixed or per effort), for the load-time warning."""
+        return self.thinking_budget or max((self.effort_budgets or {}).values(), default=None)
+
     @staticmethod
     def _strict_thinking_warning(thinking_budget: Optional[int], engine_kwargs: dict) -> Optional[str]:
         """sglang only enforces the budget with enable_strict_thinking on; without
@@ -389,9 +393,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             print(f"engine_args error: {err}", file=sys.stderr)
             return backend_pb2.Result(success=False, message=str(err))
 
-        warning = self._strict_thinking_warning(
-            self.thinking_budget or max((self.effort_budgets or {}).values(), default=None), engine_kwargs
-        )
+        warning = self._strict_thinking_warning(self._configured_budget(), engine_kwargs)
         if warning:
             print(warning, file=sys.stderr)
 

backend/python/sglang/test.py

@@ -448,6 +448,16 @@ class TestSglangHelpers(unittest.TestCase):
         self.assertIsNone(servicer._strict_thinking_warning(512, {"enable_strict_thinking": True}))
         self.assertIsNone(servicer._strict_thinking_warning(None, {}))
 
+    def test_warns_when_only_effort_budgets_are_set(self):
+        """A model with effort_budgets but no thinking_budget still needs strict thinking."""
+        servicer = self._servicer()
+        servicer.thinking_budget = None
+        servicer.effort_budgets = {"low": 256, "max": 16384}
+        self.assertEqual(servicer._configured_budget(), 16384)
+        self.assertIn("enable_strict_thinking", servicer._strict_thinking_warning(servicer._configured_budget(), {}))
+        servicer.effort_budgets = {}
+        self.assertIsNone(servicer._configured_budget())
+
     def test_explicit_zero_temperature_and_seed_are_preserved(self):
         """Temperature=0 is greedy decoding and 0 is a valid seed — neither is
         an unset value. A dropped seed turns a reproducible request random."""