feat!: Add support for Log Analytics and Remove BigQuery log destination

[daniel-cit]

This PR adds support for Log Analytics in the Log Bucket destination and removes the BigQuery log destination from the centralized logging configured in step 1-org.

Now that Log Analytics is available in GA, the same use cases for analytics that were previously accomplished using a log sink export to BigQuery can now be accomplished with a log sink export to a Logging bucket.

This change helps address use cases for both writing custom SQL queries for deep analysis analysis and ad-hoc investigation using the Logs Explorer console, with one single log sink and data source instead of two.

For more information on the differences of these log sink types see Moving to Log Analytics for BigQuery export users

[prateek2408]

Added some comments.

[jasonbisson]

The only concern I had the change would trigger a removal of the BQ dataset, but I only saw the change to the log filter.

[daniel-cit]

The only concern I had the change would trigger a removal of the BQ dataset, but I only saw the change to the log filter.

@jasonbisson
the new recommendation is to use log analytics with the log Bucket creating a Big query dataset associated with the log bucket.

This change removed the direct export to a BigQuery dataset that existed previously from the main code and from the inner module we use.

If this new version would be applied over a deployed version of the master branch it would remove an existing BigQuery dataset.

This is the reason for the feat!: to indicate a braking change in the code.

There is no in-place upgrade path from the a version of the foundation to a new version .

[gtsorbo]

LGTM

[daniel-cit]

@gtsorbo @apeabody @bharathkkb build is green :)