Support change stream filtering using MongoDB's aggregation pipeline step

Which use case/requirement will be addressed by the proposed feature?

MongoDB Kafka Source Connector's pipeline property, which allows a user to configure:

An array of aggregation pipelines to run in your change stream.

For example:

[{"$match": { "$and": [{"operationType": "insert"}, {"fullDocument.eventId": 1404 }] } }]

This feature request is to support this capability in the Debezium MongoDB connector for both change stream and snapshot modes. This would also help to give parity with the Debezium Postgres connector's additional-condition configuration parameter for ad-hoc snapshots.

Implementation ideas (optional)

This has a dependency on https://issues.redhat.com/browse/DBZ-5102.