Document skip_optimize model config for dbt-databricks

[moomindani]

What are you changing in this pull request and why?

dbt-databricks 1.12.2 adds a new skip_optimize model config that lets you opt a model out of the post-materialization OPTIMIZE call while keeping zorder / liquid_clustered_by / auto_liquid_cluster in the table definition. This documents it on the Databricks configurations page.

It complements the existing run-wide DATABRICKS_SKIP_OPTIMIZE variable by allowing per-model (and, via config inheritance, per-folder / per-project) opt-out. The added note also clarifies the precedence: DATABRICKS_SKIP_OPTIMIZE=true skips OPTIMIZE for every model and cannot be re-enabled for an individual model with skip_optimize: false.

Related: dbt-databricks PR databricks/dbt-databricks#1485, issue databricks/dbt-databricks#703

Checklist

• The changes in this PR meet the docs style guide/fundamentals required for all content.
• Applied the proper versioning rules ("Available since dbt-databricks 1.12.2" noted inline).

[vercel]

@moomindani is attempting to deploy a commit to the dbt-labs Team on Vercel.

A member of the Team first needs to authorize it.