* ci: Publish cannon prestates on develop and op-program releases.

* ci: Use new context

* ci: Remove comma

* fix branch name

* ci: Build all cannon prestates so they can be published

* ci: Not mt64 its just mt

* ci: Remove temporary settings

---------
Co-authored-by: raffaele-oplabs <raffaele@oplabs.co>

op-supervisor: return typed pairs of L1/L2 blocks, and support invalidation of fromda entries (#13840)

* op-supervisor: return typed pairs of L1/L2 blocks, and support invalidation of fromda entries

* op-supervisor: support DA db block-invalidation and replacement

* op-supervisor: fix and cover edge-case of replacing a block at later L1 scope

* op-node/rollup/derive: Reorganize baseBatchStage

Fixes #12567

* reword comment
Co-authored-by: George Knee <georgeknee@googlemail.com>

---------
Co-authored-by: George Knee <georgeknee@googlemail.com>

* op-deployer: Improve default hardfork choice

The deploy config was previously created with Holocene activated by default. This PR updates the hardfork selection logic to choose default hardforks based on the contracts version.

* CR updates

* docker: Fix RedundantTargetPlatform warnings

* docker: Fix LegacyKeyValueFormat warnings

This will host the code that interacts with:
- the definition of a devnet (including generation of kurtosis inputs)
- an arbitrary devnet, regardless of whether it's kurtosis-backed or not

On the other hand, kurtosis-devnet remains focused solely on the
instantiation of a devnet on kurtosis.

As such, kurtosis-devnet can depend on devnet-sdk, but not the other
way around.

* bump cci ci-builder version to v0.56.0

With new versions of foundry tools

* ops: update version of cffi dep

* op-e2e: Add action test to confirm fault proof program stops when the agreed timestamp is already at the game timestamp

* op-e2e: Add action tests for the challenger super trace provider

Ensures that it generates super root/transition states that match the test expectations.
Update RootProvider interface to match the actual interface exposed by Supervisor

* op-e2e: Remove game inputs that are no longer required.

* mise.toml: cleanup unused devnet tool dependencies

* op-supervisor: update devnet reference in readme

* refactor: using slices.Contains to simplify the code
Signed-off-by: pinglanlu <pinglanlu@outlook.com>

* op-batcher: reintroduce helper func to fix other usages

---------
Signed-off-by: pinglanlu <pinglanlu@outlook.com>
Co-authored-by: protolambda <proto@protolambda.com>

Signed-off-by: piguagua <piguagua@aliyun.com>

This change adds a "compilation" phase for devnet manifest.
Bottom line, we:

- treat the devnet manifest as a high-level specification of an
  expected deployment

- open up the possibility to generate an actionable input for an
  arbitrary deployer

- use kurtosis as a first target, by generating a valid kurtosis input
  that reflects the properties defined in the manifest

pkg/devnet/cmd/main.go is a toy tool to perform that last step.

Going forward, we could imagine a k8s deployment being "compiled"
following a similar process.
Or this being used during automated tests setup in order to create the
right target environment.

At a high-level, this is part of an effort to standardize our sources
of truth across the board.

Bumps [github.com/hashicorp/raft-boltdb/v2](https://github.com/hashicorp/raft-boltdb) from 2.3.0 to 2.3.1.
- [Release notes](https://github.com/hashicorp/raft-boltdb/releases)
- [Commits](https://github.com/hashicorp/raft-boltdb/compare/v2.3.0...v2.3.1)

---
updated-dependencies:
- dependency-name: github.com/hashicorp/raft-boltdb/v2
  dependency-type: direct:production
  update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* feat: Implement OPCM upgrade

* feat: better comment

* feat: Move vars to Setup where they belong

* gas snapshot

* feat: add OP_SUCCINCT game type

* fix: change OP_SUCCINCT to 5

* add ASTERISC_KONA

* op-challenger: Implement split adapter and trace accessor creation for super cannon

* op-challenger: Implement register task for super cannon

* op-challenger: Actually register super cannon game handler

* op-challenger: Lazy create clients

Make sync validator configurable per game type.

* op-challenger: Test split adapter

Remove ClaimTimestamp from ClaimInfo as we always use the poststate timestamp

Make sync validator configurable per game type.

* feat(op-challenger): Kona interop executor

* Add agreed prestate to `LocalGameInputs`

* super executor 🦸

* lint

Not currently used in the interop devnet as we're not yet deploying the new game type.

This is a convenience feature that allows a user to "enter" a devnet /
chain and get a subshell with relevant data (in particular env
variables).

The goal is to try an minimize the amount of error-prone log-browsing
and copy-pasting before being able to interact with the environment.

* op-chain-ops: op-run-block debug tool

* op-run-block: fix conflict with geth merge

* op-program: Implement retries in prefetcher block re-exec

* add canon block with error test case

* add comment on error case in rexec

* op-program: Implement consolidation for interop

* address todos; fix claim validation bug

* fix test

* fix todos

* op-program: Fix short-circuit logic in block walk

* op-program: Use l2-chain block fetcher in canon

* op-program: Fix current header in engine_backend

* op-program: Filter hazardrous exec msgs

* cleanup

* fix block cache lookup

* snap to earliest block and update tests
Co-authored-by: Adrian Sutton <adrian@oplabs.co>

* use adrian's suggestion

* todo formatting

---------
Co-authored-by: Adrian Sutton <adrian@oplabs.co>