test: bump TestBlockchain.DefaultTimeout 10s -> 30s

[LukaszRozmej]

Summary

• executePayloadV1_accepts_already_known_block(true) flakes on TEST_USE_FLAT=1 runs with TaskCanceledException at exactly 10s 489ms, inside TestBlockchain.Build() while waiting for the NewHeadBlock event after genesis loads.
• The cancellation token used there is this.CancellationToken, which derives from TestTimeout (default 10000), not the test's [CancelAfter(30000)].
• PR test: bump 10s timeouts that flake on Flat DB CI #11380 bumped per-test [CancelAfter] values from 10s/5s to 30s but missed this internal one. Aligning DefaultTimeout with that fix.

Stack from the failing run:

System.Threading.Tasks.TaskCanceledException : A task was canceled.
   at Nethermind.Core.Events.Wait.ForEventCondition[T](...) in Nethermind.Core/Events/Wait.cs:33
   at Nethermind.Core.Test.Blockchain.TestBlockchain.Build(...) in TestBlockchain.cs:227
   at MergeTestBlockchain.Build(ISpecProvider) in BaseEngineModuleTests.cs:292
   at EngineModuleTests.executePayloadV1_accepts_already_known_block(...) in EngineModuleTests.V1.cs:464

Side effects of the bump:

• BlockProcessorTests.cs:146 uses DefaultTimeout * 10 — 100s ceiling becomes 300s. Just a higher upper bound; passing tests don't notice.
• TestRpcBlockchain.cs:72 uses it as a fallback when no explicit timeout is passed.

Test plan

• CI green on Nethermind tests (Flat DB) — specifically Nethermind.Merge.Plugin.Test.
• No regressions in existing tests that rely on the timeout.

🤖 Generated with Claude Code

[claude]

Claude finished @LukaszRozmej's task in 1m 18s —— View job

---

Review

• Read changed file and related files
• Check side effects mentioned in PR description
• Review test infrastructure rules
• Post final review

---

Change: DefaultTimeout = 10000 → DefaultTimeout = 30000 in TestBlockchain.cs:49

The fix is correct and well-targeted. The root cause is clear: TestBlockchain.Build() at line 225 uses this.CancellationToken, which is sourced from CreateCancellationSource() → AutoCancelTokenSource.ThatCancelAfter(TestTimeout) (line 390), where TestTimeout initializes from DefaultTimeout. This is a separate timeout budget from the NUnit [CancelAfter(30000)] on the test method itself — so bumping the NUnit attribute in PR #11380 left this internal one at 10s, still too tight for Flat DB genesis initialization.

Side effects verified:

• BlockProcessorTests.cs:146 — DefaultTimeout * 10 becomes 300s (was 100s). This is a guard ceiling for SemaphoreSlim.WaitAsync; a passing test will never approach it. No concern.
• TestRpcBlockchain.cs:72 — used as the fallback TestTimeout when callers pass no explicit timeout. Raising from 10s to 30s is appropriate given the same Flat DB slowdowns apply to RPC blockchain tests.

No issues found. The change is minimal, focused, and consistent with the intent of PR #11380. All known consumers of DefaultTimeout are accounted for in the PR description.

Verdict: approve.