fix: Batching failures when individual event errors bubble up

[bmtcril]

When errors such as when a backend isn't configured or there is no transformer for an event, bubbles up it causes the whole batch to get dumped to the dead letter queue. These changes cause those errors to be logged but otherwise just get dropped from the batch.

Error:

[2024-03-27 14:58:10,291: INFO/ForkPoolWorker-23] Event edx.course.enrollment.deactivated has been queued for batching. Queue size: 10
[2024-03-27 14:58:10,291: INFO/MainProcess] Task event_routing_backends.tasks.dispatch_bulk_events[c6ff920d-3d73-410d-ab73-bd41cfae0ec1] received
[2024-03-27 14:58:10,292: ERROR/ForkPoolWorker-23] Exception occurred while trying to bulk dispatch 10 events.
Traceback (most recent call last):
  File "/openedx/venv/lib/python3.8/site-packages/event_routing_backends/backends/events_router.py", line 187, in send
    self.bulk_send([json.loads(queued_event.decode('utf-8')) for queued_event in batch])
  File "/openedx/venv/lib/python3.8/site-packages/event_routing_backends/backends/events_router.py", line 153, in bulk_send
    event_routes = self.prepare_to_send(events)
  File "/openedx/venv/lib/python3.8/site-packages/event_routing_backends/backends/events_router.py", line 93, in prepare_to_send
    processed_events = self.process_event(event)
  File "/openedx/venv/lib/python3.8/site-packages/event_routing_backends/backends/events_router.py", line 256, in process_event
    events = processor(events)
  File "/openedx/venv/lib/python3.8/site-packages/event_routing_backends/processors/mixins/base_transformer_processor.py", line 37, in __call__
    transformed_event = self.transform_event(event)
  File "/openedx/venv/lib/python3.8/site-packages/event_routing_backends/processors/caliper/transformer_processor.py", line 43, in transform_event
    raise NoBackendEnabled
eventtracking.processors.exceptions.NoBackendEnabled
[2024-03-27 14:58:10,292: INFO/ForkPoolWorker-23] Pushing failed events to the dead queue: dead_queue_caliper

[Ian2012]

great change here