fix: add missing provider aliases, fix test semaphore accounting

- Add vertex_ai_anthropic → anthropic, bedrock_converse → aws,
  azure_ai → azure to _PROVIDER_MAP so composite providers report
  the canonical slug instead of their raw internal key
- Fix test_fire_and_forget_respects_semaphore_cap to release only
  as many slots as were actually acquired, preventing semaphore
  count from exceeding _MAX_INFLIGHT if an assert fires mid-loop

litellm/integrations/tickerr.py

@@ -48,7 +48,10 @@ _PROVIDER_MAP: Dict[str, str] = {
     "fireworks_ai": "fireworks",
     "openrouter": "openrouter",
     "azure": "azure",
+    "azure_ai": "azure",
     "bedrock": "aws",
+    "bedrock_converse": "aws",
+    "vertex_ai_anthropic": "anthropic",
     "ai21": "ai21",
     "cerebras": "cerebras",
     "xai": "xai",

tests/logging_callback_tests/test_tickerr_callback.py

@@ -217,10 +217,21 @@ def test_report_omits_error_type_for_unknown_code():
 
 def test_fire_and_forget_respects_semaphore_cap():
     """Reports beyond _MAX_INFLIGHT are dropped silently without blocking."""
-    # Exhaust the semaphore by acquiring all slots directly
+    # Exhaust the semaphore by acquiring all slots directly.
+    # Track how many we actually acquired so the finally block releases exactly
+    # that many — releasing more than acquired would push the count above its
+    # initial maximum and corrupt later tests.
+    acquired_count = 0
     for _ in range(_MAX_INFLIGHT):
-        acquired = _inflight.acquire(blocking=False)
-        assert acquired, "semaphore should have slots available at test start"
+        if _inflight.acquire(blocking=False):
+            acquired_count += 1
+        else:
+            break
+
+    assert acquired_count == _MAX_INFLIGHT, (
+        f"semaphore should have {_MAX_INFLIGHT} slots available at test start, "
+        f"got {acquired_count}"
+    )
 
     try:
         # With semaphore exhausted, _fire_and_forget must return immediately
@@ -229,7 +240,7 @@ def test_fire_and_forget_respects_semaphore_cap():
             _fire_and_forget({"provider": "openai"})
             mock_thread.assert_not_called()
     finally:
-        for _ in range(_MAX_INFLIGHT):
+        for _ in range(acquired_count):
             _inflight.release()