docs: pre-aggregation data source activation, variable families and driver_factory precedence

[igorlukanin]

Follow-up to #10587. The main documentation for pre-aggregation-specific data source credentials landed in 3a84d1a, but a few points that trip readers up were missing.

• Activation rule. Setting any one credential variable with the PRE_AGGREGATIONS segment switches the data source onto the dedicated connection — there is no separate flag. Called out explicitly, along with the four CUBEJS_PRE_AGGREGATIONS_* variables that share the prefix but configure behavior rather than a connection (SCHEMA, BUILDER, BACKOFF_MAX_TIME, ALLOW_NON_STRICT_DATE_RANGE_MATCH) and so deliberately do not trigger it. These look like they would, which was the most confusing part.
• Variable families. The segment isn't limited to _DB_* — JDBC_, AWS_, DATABASE and FIREBOLT_ take it too. Previously an Athena or Firebolt user would conclude the feature didn't apply to them.
• driver_factory precedence. A custom driver_factory wins over the env vars and logs a Pre-aggregation driver conflict warning. This was undocumented on both the pre-aggregations page and the driver_factory reference.

Also replaced the N/A placeholders in the CUBEJS_PRE_AGGREGATIONS_DB_* reference table with Not set.

yarn broken-links --check-anchors reports no issues in the changed files (the 16 pre-existing failures are all in embedding/iframe/, untouched here).

[mintlify]

Preview deployment for your docs. Learn more about Mintlify Previews.

Project	Status	Preview	Updated (UTC)
cubed3	🟢 Ready	View Preview	Jul 28, 2026, 3:15 AM

💡 Tip: Enable Workflows to automatically generate PRs for you.