Support hashing of search document ID

[eminano]

Currently the search adapter supports using the primary key as the document ID. It does so by prepending the value of the primary column (or combination of values in the case of a composite primary key) by the table pgstream id.
However, depending on the value of the primary key column, this can get large quickly. Opensearch currently has a size limit of the document ID field of 512 bytes. In order to ensure this limit is not reached, we could hash the primary key value(s) to ensure the limit is not reached.

The ideal solution would be to add an option to the library where the user can provide the hashing mechanism of their choosing, so that the same can be used when querying the data. By default, we'll expose a hashing mechanism that will be used when using the CLI as a default when hashing is selected.