Add batching controls to `MessageQueueTrigger`

### Description

I would like to propose adding batching controls to `MessageQueueTrigger`.

Currently, a DAG triggered by a messaging-based asset may receive multiple events in `triggering_asset_events`, but this does not seem to be configurable in a general way at the `MessageQueueTrigger` level, even if some providers may support provider-specific behavior.

It would be useful to support, for example, a parameter to control the maximum batch size and another one to control the maximum wait time before flushing.

### Use case/motivation

This would make `MessageQueueTrigger` easier to use in cases where message batching needs to be bounded and more predictable.

### Related issues

_No response_

### Are you willing to submit a PR?

- [x] Yes I am willing to submit a PR!

### Code of Conduct

- [x] I agree to follow this project's [Code of Conduct](https://github.com/apache/airflow/blob/main/CODE_OF_CONDUCT.md)

Start by locating the Python implementation of MessageQueueTrigger and tracing how messaging-based assets populate triggering_asset_events. Compare any provider-specific batching behavior, then define how configurable maximum batch size and maximum wait time should apply; done means both controls are supported consistently at the trigger level.

Written by the indexing model from the issue text.