Respect file extension setting for S3 destinations in Firehose provider

The Firehose delivery stream implementation was not considering the file
extension setting when the target was defined as an S3 bucket. This fix
ensures that the specified file extension is appended to the S3 object's
key when writing data. The `_get_s3_object_path` method now accepts the
`file_extension` parameter to accommodate this enhancement.

localstack/services/firehose/provider.py

@@ -765,6 +765,7 @@ def _put_records_to_s3_bucket(
  ):
  bucket = s3_bucket_name(s3_destination_description["BucketARN"])
  prefix = s3_destination_description.get("Prefix", "")
+ file_extension = s3_destination_description.get("FileExtension", "")
 
  if role_arn := s3_destination_description.get("RoleARN"):
  factory = connect_to.with_assumed_role(
@@ -777,15 +778,15 @@ def _put_records_to_s3_bucket(
  )
  batched_data = b"".join([base64.b64decode(r.get("Data") or r.get("data")) for r in records])
 
- obj_path = self._get_s3_object_path(stream_name, prefix)
+ obj_path = self._get_s3_object_path(stream_name, prefix, file_extension)
  try:
  LOG.debug("Publishing to S3 destination: %s. Data: %s", bucket, batched_data)
  s3.put_object(Bucket=bucket, Key=obj_path, Body=batched_data)
  except Exception as e:
  LOG.exception(f"Unable to put records {records} to s3 bucket.")
  raise e
 
- def _get_s3_object_path(self, stream_name, prefix):
+ def _get_s3_object_path(self, stream_name, prefix, file_extension):
  # See https://aws.amazon.com/kinesis/data-firehose/faqs/#Data_delivery
  # Path prefix pattern: myApp/YYYY/MM/DD/HH/
  # Object name pattern: DeliveryStreamName-DeliveryStreamVersion-YYYY-MM-DD-HH-MM-SS-RandomString
@@ -794,6 +795,10 @@ def _get_s3_object_path(self, stream_name, prefix):
  pattern = "{pre}%Y/%m/%d/%H/{name}-%Y-%m-%d-%H-%M-%S-{rand}"
  path = pattern.format(pre=prefix, name=stream_name, rand=str(uuid.uuid4()))
  path = timestamp(format=path)
+
+ if file_extension:
+ path += file_extension
+
  return path
 
  def _put_to_redshift(