Customize final destination schemas/datasets when configuring a source

[ChristopheDuong]

Tell us about the problem you're trying to solve

• Currently, the schema (or dataset for BigQuery) is chosen when creating the destination. Up until now, we expected the data to be replicated and written in that destination "folder"
• Following the work on Change jdbc sources to discover more than standard schemas #1038 we can now replicate multiple schemas at once when setting up a single DB source.
• However this could fail especially in multi-tenants scenarios, we could find conflicts in table naming from different schemas. The user would potentially like to keep tables from different schemas separated or merge them all together.
• Or as noted in Prefix table names in destinations with the source name #973, different sources could also use generic table names such as users which would definitely not be merge-able with different schemas without having to prefix table names with the source type.
• Another concern is that it is sometimes desirable to not touch table names (adding prefix) but only tweak schemas .

Describe the solution you’d like

In the UI, when selecting the tables to be replicated when configuring a Catalog, it would be helpful to override the default schema defined in the destination per stream.

A new field in the Configured Catalog could be created to handle such customizable "namespaces" and eventual table renaming.

Such feature has been discussed and requested by users such as:

• https://airbytehq.slack.com/archives/C01AHCDHCKE/p1605804939047900?thread_ts=1605795947.045500&cid=C01AHCDHCKE
• Naming conventions managed in destinations #1060 (comment)