Improve test timeout diagnostics with crash reports and stack traces
When a test times out, the system now collects crash reports from Redis servers and stack traces from clients before terminating, providing better diagnostics for hung tests instead of silently killing them.