Destination BigQuery: add the ability to partition by _airbyte_extracted_at

[KimPlv]

Hi,
Is it possible to give the option of selecting the partition field between _airbyte_generation_id and _airbyte_extracted_at for Bigquery tables. It seems more optimized to reverse and have _airbyte_extracted_at for the partition and _airbyte_generation_id for clustering. (Btw why did you choose this configuration? ) But at least, if we can have the possibility of having the old version of the partition as an option, that would be great.

[tmas-definitive]

I think this is a great idea. I have a very large source which never requires a refresh of data (hence _airbyte_generation_id never increments) and I end up having huge compute costs since it is not partitioned on _airbyte_extracted_at. Partitioning on extracted_at would be 2 orders of magnitude more efficient in the jobs bytes processed in my case