supervisor: L1 Finality Processor

[axelKingsley]

What

Extends the L1 Processor from #13206 to include subscription and management of the L1 Finality Signal

Why

We would like for the Supervisor to control the canonical L1 source for "Owned Nodes" in order to have better consistency and control over the data in the system.

How

Introduces:

• Eth Subscription to finality
• Update of L1 Finality to the ChainsDB on handle
• Stopping L1 Finality updates from OP Nodes

Testing

• Existing tests and checks.
• New unit tests for basic update/don't-update behavior of the handler
• Additionally Tested by running the existing E2E tests with a long wait-time and observing the logs:

    l1_processor.go:109:        WARN [12-05|15:47:50.851] Failed to process L1                     role=supervisor service=l1-processor err="failed to fetch header by num 11: not found"
    l1_processor.go:106:        DEBUG[12-05|15:47:56.838] Checking for new L1 block                role=supervisor service=l1-processor current=10
    l1_processor.go:147:        DEBUG[12-05|15:47:56.838] Received new Finalized L1 block          role=supervisor service=l1-processor block=81e1e0..d7c67e:8
    l1_processor.go:160:        DEBUG[12-05|15:47:56.838] Updated finalized L1 block               role=supervisor service=l1-processor block=81e1e0..d7c67e:8
    l1_processor.go:128:        DEBUG[12-05|15:47:56.838] Processing new L1 block                  role=supervisor service=l1-processor block=8dfb55..df45d5:11
    l1_processor.go:106:        DEBUG[12-05|15:47:56.850] Checking for new L1 block                role=supervisor service=l1-processor current=11
    l1_processor.go:109:        WARN [12-05|15:47:56.850] Failed to process L1                     role=supervisor service=l1-processor err="failed to fetch header by num 12: not found"
    l1_processor.go:147:        DEBUG[12-05|15:47:56.850] Received new Finalized L1 block          role=supervisor service=l1-processor block=81e1e0..d7c67e:8
    l1_processor.go:106:        DEBUG[12-05|15:48:02.838] Checking for new L1 block                role=supervisor service=l1-processor current=11
    l1_processor.go:147:        DEBUG[12-05|15:48:02.839] Received new Finalized L1 block          role=supervisor service=l1-processor block=5d5aa2..305c6f:9
    l1_processor.go:160:        DEBUG[12-05|15:48:02.839] Updated finalized L1 block               role=supervisor service=l1-processor block=5d5aa2..305c6f:9

When the OP Nodes were still contributing their L1 Finalized, this function was very rarely able to get an update it. It would discover one, and then find it already set in the database.

[axelKingsley]

did a brief review with @protolambda synchronously today:

• To fix the action tests, directly leverage the L1 Processor's handleFinality function (expose it)

[codecov]

Codecov Report

Attention: Patch coverage is 42.10526% with 33 lines in your changes missing coverage. Please review.

Project coverage is 42.95%. Comparing base (eabf704) to head (1b8e22c).
Report is 14 commits behind head on develop.

Files with missing lines	Patch %	Lines
...isor/supervisor/backend/processors/l1_processor.go	55.00%	17 Missing and 1 partial ⚠️
op-service/sources/supervisor_client.go	0.00%	7 Missing ⚠️
op-supervisor/supervisor/backend/backend.go	0.00%	4 Missing ⚠️
op-supervisor/supervisor/backend/db/query.go	0.00%	2 Missing ⚠️
op-supervisor/supervisor/backend/mock.go	0.00%	2 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop   #13274      +/-   ##
===========================================
- Coverage    47.45%   42.95%   -4.50%     
===========================================
  Files          931      762     -169     
  Lines        78188    68462    -9726     
  Branches       849        0     -849     
===========================================
- Hits         37103    29408    -7695     
+ Misses       38361    36554    -1807     
+ Partials      2724     2500     -224     

Flag	Coverage Δ
cannon-go-tests-32	?
cannon-go-tests-64	?
contracts-bedrock-tests	?

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

Files with missing lines	Coverage Δ
op-node/rollup/interop/interop.go	56.17% <ø> (+0.20%)	⬆️
op-service/testutils/mock_interop_backend.go	62.74% <ø> (-7.07%)	⬇️
op-supervisor/supervisor/frontend/frontend.go	100.00% <100.00%> (ø)
op-supervisor/supervisor/backend/db/query.go	0.00% <0.00%> (ø)
op-supervisor/supervisor/backend/mock.go	0.00% <0.00%> (ø)
op-supervisor/supervisor/backend/backend.go	35.59% <0.00%> (-0.21%)	⬇️
op-service/sources/supervisor_client.go	0.00% <0.00%> (ø)
...isor/supervisor/backend/processors/l1_processor.go	54.45% <55.00%> (-0.24%)	⬇️

... and 178 files with indirect coverage changes

[axelKingsley]

Sorry for the thrash -- I rebased this PR in a way that caused the new commits to become detached from the branch. Then when I pushed it cleared all the changes which auto-closed the PR. I've fixed it and squashed the changes into one commit.

[protolambda]

LGTM, just one type change needed