Feature request or enhancement

Add automatic retry to Debezium Vitess Connector snapshots.

In DBZ-5930 snapshot support was added without automatic retry. There have been changes/fixes in upstream vitess to make the vgtid an opaque state representation of how far along a snapshot is (e.g., https://github.com/vitessio/vitess/pull/12623). We can parse and use this vgtid information in order to allow for automatic retry of a snapshot upon failure.

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

This is a crucial feature for large production tables where snapshots will take a long time and we cannot guarantee the tasks will be able to run for the entirety of the snapshot. With the current logic there is not auto-retry or failure recovery. Meaning users have to start again from the beginning which can be very expensive/time-consuming.

Implementation ideas (optional)

Will put out a PR