op-node/rollup/interop: do not ignore local unsafe reset

[nonsense]

Description

This PR is:

1. Fixing a bug with L2 unsafe head not reorging after a L1 reorg
2. Fixing a config with deployment of interop network in sysgo tests -- SequencerConfDepth
3. Fixing the wait times of a few tests due to the change in 2.

---

In managed-mode in interop networks, it is the op-supervisor that monitors the L1 chain, and is the source-of-truth for the networks view for the op-node. At the moment when it issues a reset, the op-node is ignoring the LocalUnsafe field within the reset event.

This means that if the L2 chain has progressed on an incorrect chain due to L1 reorg, its unsafe ref is not rewound back to a block using a valid L1 ref after the reorg.

Tests

This functionality is tested with TestL2ReorgAfterL1Reorg from #16105

cd op-acceptance-tests/tests/interop/reorgs
go test -count=1 -timeout=30m -failfast -v -run TestL2ReorgAfterL1Reorg ./...

If you comment out the update to LocalUnsafe field, the test will fail, and you can inspect the logs and see that L1Origin ref for the L2 chain blocks is stuck on a reorged L1 block.

I haven't added the test to this PR, as the op-test-sequencer L1-mode needs more work in order to be ready for review, but this bugfix and the test itself are ready.

• add test to this PR to visualize and confirm this bug - test added at op-test-sequencer: support for L1 reorgs #16105 to keep change-sets short and easy to review.

Follow-up

In a separate PR, as a follow-up:

• improve op-test-sequencer so that we can run the test multiple times. when running with -count=2, no reorg on L1 is triggered at the moment
• use op-supervisor as source-of-truth for L1 origin selector (op-node.rollup.sequencing.L1OriginSelector)

Additional context

Found as part of #15331

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 96.24%. Comparing base (dc1cca7) to head (957c538).
Report is 5 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff              @@
##           develop   #16052       +/-   ##
============================================
+ Coverage    81.55%   96.24%   +14.68%     
============================================
  Files          161      106       -55     
  Lines         8866     4583     -4283     
============================================
- Hits          7231     4411     -2820     
+ Misses        1502      172     -1330     
+ Partials       133        0      -133     

Flag	Coverage Δ
cannon-go-tests-64	?
contracts-bedrock-tests	96.24% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

see 55 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[Copilot]

Pull Request Overview

This PR fixes an issue in managed-mode interop where the op-node was ignoring the local unsafe field during a reset, which could lead to inconsistencies after L1 reorganizations.

• Update the Reset method to retrieve the latest valid local unsafe block instead of using an empty reference.
• Introduce a new function (scanL2ForLatestLocalUnsafe) to scan the L2 chain and validate the L1 origins.
• Add new methods to the L2Source and L1Source interfaces to support block retrieval by label and by number.

[nonsense]

Due to SequencerConfDepth: 2 change, a few tests' waits had to be bumped up with at least 2*blockTime (12sec. in tests).