Motivation

Waiting for the dhandle sweep server in the Python suite is done several ways with inconsistent semantics:

• test_layered_follower14 has a bounded wait_for_sweep() (120x1s) that asserts on timeout.
• test_sweep03 has a bounded wait_for_sweep(baseline) (10x0.5s) that silently falls through on timeout (returns a count, no assertion).
• test_layered_checkpoint08 polls dh_sweep_dead_close directly (bounded in WT-17719).

These should be unified into one reusable helper with consistent, assert-on-timeout behaviour so sweep-wait loops fail fast with a clear message instead of hanging to a task-level timeout.

Approach

• Add a reusable bounded wait_for_sweep(self, baseline, min_sweeps=1, timeout=60) helper that asserts with a clear message on timeout. Sweep waiting is not disaggregated-specific (test_sweep* use it too), so place it on WiredTigerTestCase in test/suite/helpers/wttest.py.
• Migrate the existing wait_for_sweep copies and inline sweep-stat poll loops to the shared helper.

Definition of done

• Shared bounded wait_for_sweep helper added.
• Existing copies/inline loops migrated, standardised on assert-on-timeout semantics.
• Tests still pass on the happy path.

Follow-up to WT-17719 (see also WT-17700).