[Connector] Add bound-parameter support to databricks/execute_sql

[michaelmcnees]

Background

The Databricks SQL Statements API supports named parameters — callers can pass {{param_name}} placeholders in the SQL and a parameters array in the request body:

{
  "statement": "SELECT * FROM users WHERE id = :id",
  "parameters": [{"name": "id", "value": "42", "type": "INT"}]
}

This is different from the ?/$N placeholder approach used by the other SQL connectors (MySQL, MSSQL, Redshift, Snowflake) which pass args via database/sql's variadic QueryContext.

Proposed change

Add an optional parameters param to DatabricksExecuteSQLConnector.Execute:

- action: databricks/execute_sql
  params:
    warehouse_id: "..."
    statement: "SELECT * FROM orders WHERE user_id = :user_id AND status = :status"
    parameters:
      - name: user_id
        value: "{{ inputs.user_id }}"
        type: STRING
      - name: status
        value: active
        type: STRING

The parameters array should be passed through to the Databricks API payload as-is (after type-asserting each entry as map[string]any). No driver-level escaping is needed — Databricks handles parameterization server-side.

Files

• packages/engine/internal/connector/databricks.go — DatabricksExecuteSQLConnector.Execute
• packages/engine/internal/connector/databricks_test.go — add a test asserting the parameters field is included in the request body

References

• Deferred during PR feat(connector): enterprise and cloud connector batch (batch 3) #149 review (CodeRabbit flagged databricks/execute_sql as needing bound-parameter parity with the other SQL connectors)
• Databricks SQL Statements API: https://docs.databricks.com/api/workspace/statementexecution/executestatement