Test Suite Speedup — Design

Goal

Reduce full test suite from ~82s to ~55s by fixing the 5 slowest tests (~25s total waste) without changing the behavior they validate.

Problem

5 tests account for ~25s of the 82s total:

Test	Time	Root Cause
3 orchestration integration tests	~5s each	CPU checker: psutil.cpu_percent(interval=1.0) × 5 samples
1 disk scanning test	~5s	Real filesystem scan of /tmp via du + Path.rglob
1 timeout test	~5s	Intentional time.sleep(5) in test thread

Fixes

Fix 1: Mock CPU sampling in 3 integration tests (~15s saved)

Tests: test_pipeline_with_optional_failing_node, test_transform_between_nodes, test_context_to_intelligence_pipeline in apps/orchestration/_tests/test_integration.py.

These test pipeline orchestration logic, not CPU measurement accuracy. Patch psutil.cpu_percent to return a fixed value instantly.

Fix 2: Reduce sleep in timeout test (~5s saved)

Test: test_timeout_branch in apps/orchestration/_tests/test_nodes.py.

Change time.sleep(5) to time.sleep(0.5) — still well above the 0.01s timeout, proves the same behavior 10x faster.

Fix 3: Mock filesystem in disk scan test (~5s saved)

Test: test_provider_disk_progress_callback in apps/intelligence/_tests/providers/test_local.py.

Patch subprocess.run and/or Path.rglob to return fixture data, or use a small temp directory instead of scanning all of /tmp.

Expected Result

Before	After	Saved
~82s	~55s	~27s (33%)

Out of Scope

• Admin test optimization (inherent DB/rendering costs, diminishing returns)
• pytest-xdist parallelization (separate concern, can be added later)
• Pre-commit hook changes (already fast at ~6s with black + ruff + mypy)