revert(metrics): typing regression on log_metrics callable (aws-powertools#744)

Author: heitorlessa

aws_lambda_powertools/metrics/metrics.py

@@ -2,9 +2,8 @@
 import json
 import logging
 import warnings
-from typing import Any, Callable, Dict, Optional, Union, cast
+from typing import Any, Callable, Dict, Optional, Union
 
-from ..shared.types import AnyCallableT
 from .base import MetricManager, MetricUnit
 from .metric import single_metric
 
@@ -130,7 +129,7 @@ def log_metrics(
         capture_cold_start_metric: bool = False,
         raise_on_empty_metrics: bool = False,
         default_dimensions: Optional[Dict[str, str]] = None,
-    ) -> AnyCallableT:
+    ):
         """Decorator to serialize and publish metrics at the end of a function execution.
 
         Be aware that the log_metrics **does call* the decorated function (e.g. lambda_handler).
@@ -170,14 +169,11 @@ def handler(event, context):
         # Return a partial function with args filled
         if lambda_handler is None:
             logger.debug("Decorator called with parameters")
-            return cast(
-                AnyCallableT,
-                functools.partial(
-                    self.log_metrics,
-                    capture_cold_start_metric=capture_cold_start_metric,
-                    raise_on_empty_metrics=raise_on_empty_metrics,
-                    default_dimensions=default_dimensions,
-                ),
+            return functools.partial(
+                self.log_metrics,
+                capture_cold_start_metric=capture_cold_start_metric,
+                raise_on_empty_metrics=raise_on_empty_metrics,
+                default_dimensions=default_dimensions,
             )
 
         @functools.wraps(lambda_handler)
@@ -198,7 +194,7 @@ def decorate(event, context):
 
             return response
 
-        return cast(AnyCallableT, decorate)
+        return decorate
 
     def __add_cold_start_metric(self, context: Any) -> None:
         """Add cold start metric and function_name dimension