create changestream with timeout

[jgao54]

Calls client.Watch with a 5 minute context deadline so the driver sends maxTimeMS over the wire which is honored by the server. Otherwise server-side default maxTimeMS is used which can sometimes be too short for the initial aggregation.

Also added better logging for when change stream is recreated.

Fixes: https://linear.app/clickhouse/issue/DBI-672

[github-actions]

🔄 Flaky Test Detected

Analysis: The e2e test suite hit the 900s timeout exactly (ran 900.557s), indicating the tests were killed by the deadline rather than failing with a specific assertion — a classic flaky timeout pattern in integration/e2e suites.
Confidence: 0.88

✅ Automatically retrying the workflow

View workflow run

[codecov]

❌ 2 Tests Failed:

Tests completed	Failed	Passed	Skipped
2165	2	2163	183

View the top 2 failed test(s) by shortest run time

github.com/PeerDB-io/peerdb/flow/e2e::TestPeerFlowE2ETestSuiteMySQL_CH

Stack Traces | 0.01s run time

=== RUN   TestPeerFlowE2ETestSuiteMySQL_CH
=== PAUSE TestPeerFlowE2ETestSuiteMySQL_CH
=== CONT  TestPeerFlowE2ETestSuiteMySQL_CH
--- FAIL: TestPeerFlowE2ETestSuiteMySQL_CH (0.01s)

github.com/PeerDB-io/peerdb/flow/e2e::TestPeerFlowE2ETestSuiteMySQL_CH/Test_Column_Exclusion

Stack Traces | 5.02s run time

=== RUN   TestPeerFlowE2ETestSuiteMySQL_CH/Test_Column_Exclusion
=== PAUSE TestPeerFlowE2ETestSuiteMySQL_CH/Test_Column_Exclusion
=== CONT  TestPeerFlowE2ETestSuiteMySQL_CH/Test_Column_Exclusion
2026/04/13 21:09:17 INFO Received AWS credentials from peer for connector: ci x-peerdb-additional-metadata={Operation:FLOW_OPERATION_UNKNOWN}
2026/04/13 21:09:17 INFO Received AWS credentials from peer for connector: clickhouse x-peerdb-additional-metadata={Operation:FLOW_OPERATION_UNKNOWN}
    clickhouse_test.go:1859: Failed to connect temporal client: failed reaching server: context deadline exceeded
--- FAIL: TestPeerFlowE2ETestSuiteMySQL_CH/Test_Column_Exclusion (5.02s)

To view more test analytics, go to the Test Analytics Dashboard
_{📋 Got 3 mins? Take this short survey to help us improve Test Analytics.}

[github-actions]

🔄 Flaky Test Detected

Analysis: Infrastructure instability during the test run — the PostgreSQL catalog was terminated mid-run (SQLSTATE 57P01) and the Temporal server became unreachable (context deadline exceeded), causing cascading test failures unrelated to any code change.
Confidence: 0.95

✅ Automatically retrying the workflow

View workflow run

[github-actions]

🔄 Flaky Test Detected

Analysis: The test TestPeerFlowE2ETestSuiteMySQL_CH/Test_Column_Exclusion failed with a context deadline exceeded error when connecting to the Temporal server, indicating a transient infrastructure/network timeout rather than a real code regression.
Confidence: 0.92

✅ Automatically retrying the workflow

View workflow run