## Problem
The pipeline creation wizard in #49243 needs destination connection
settings on its connection step, separately from replication settings
and table layout. These fields currently live inside the sheet’s
Advanced settings accordion.
## Solution
Extract BigQuery’s **Connection pool size** and **Maximum staleness**,
DuckLake’s **Pool size**, and Snowflake’s optional **Role** into
reusable connection-field components. Keep the existing accordion as
their first consumer. Labels, defaults, units, field order and
empty-value handling stay the same.
## Review instructions
1. Open the [preview to Database →
Pipelines](https://studio-staging-git-dnywh-refactorpipeline-conne-4a9108-supabase.vercel.app/dashboard/project/_/database/pipelines),
select a test project with Pipelines enabled and click **Add pipeline**.
Keep BigQuery selected and expand **Advanced settings**.
2. Confirm **Connection pool size** and **Maximum staleness** still
appear below **Invalidated slot behavior** and above **Table layout**,
with the same descriptions and the **connections** and **minutes**
units.
5. Select DuckLake, if available, and expand **Advanced settings**.
Confirm **Pool size** remains above **Batch wait time**, with its
existing description and 1–6 limits.
6. Select Snowflake, if available, and expand **Advanced settings**.
Confirm optional **Role** remains below **Invalidated slot behavior**,
with the `PIPELINES_ROLE` placeholder.
Danny White
a42433c9bd