On Mac and Windows, we should be relatively safe using
host.docker.internal (feature of Docker Desktop)

On Linux, try to find the docker bridge and infer the IP from
there.

* feat: Use encodeCall for type safety

A very common issue (both from experience and observation) is
failing to correctly encode the initializer arguments in
`encodeSystemConfigInitializer()`. This fixes that.

I believe that at one point we needed to be able to pass the 4 bytes
selector in as a string, but that no longer seems to be the case.

* semver

* feat: expand use of encodeCall to rest of OPCM

* semver lock

* cannon arch: move derivable stuff to arch.go

* address comments

This is a code cleanup change.

This moves all the code built on top of Kurtosis SDK into separate
packages for better clarity.

* cannon: Add new state version and start migrating

* cannon: Add new test state for version 4

* cannon: Update cannon commands to use new state version

* cannon: Update cannon embeds command

* cannon: Update Dockerfile to include cannon version 3 binary

- fix the op-challenger image build, as currently
  ops/docker/op-stack-go/Dockerfile defaults to a non-existing tag.

- make it easier to point to an alternate kurtosis package (like a
local for for example)

Adjust the definition, based on latest changes in kurtosis
optimism-package.

For the purpose of seeding op-challenger, we need only the proof files.

This change adds a separate target that outputs only these.

Unfortunately, docker outputs can be pretty slow, so avoiding dumping
the heavier binaries makes a significant differente performance-wise.

Now that https://github.com/ethpandaops/optimism-package/pull/116
defines a cannon_prestates_url for each challenger instance, we better
skip rebuilding the same thing multiple times.

Now that we don't use the kurtosis cli, there's no reason to go
through the filesystem to pass a configuration file.

This change reimplements run over the kurtosis SDK instead of shelling
out to the CLI tool, which fixes potential consistency problems with
other uses of the SDK.

Additionally, we split the endpoint detection logic into a separate
entity (will make it a separate package a bit later), and improve the
host handling part by no longer assuming reachability on localhost
(implicitly assuming Docker backend). This is a step toward supporting
k8s backend transparently.

* feat: remove enableLegacyContracts

* feat: remove setUseFaultProofs

* feat: Remove useFaultProofs branch in Deploy._run()

* gas snapshot

* feat: remove useFaultProofs

* Revert "op-batcher: Fix concurrent map write (#13527)"

This reverts commit 90435d07.

* lock channelMgrMutex during recordFailedDARequest

* rename / export pruneSafeBlocks and pruneChannels

mtx on channel_manager too

This prepares the ground for op-challenger needing a URL locator
serving prestate proof files.

A {{localPrestate}} placeholder will expand to a URL that's usable by
op-challenger.

* [op-program] Update releases.json

* [op-program] Tweak verify formatting

This ties together the various capabilities in pkg/ and orchestrates them. The phases are:
- build local artifacts per input template specification
- put them where they can be accessed at runtime by kurtosis
- run the deployment
- collect useful information and output it for downstream consumption

Also provide sample devnet definitions

Those are just starting points. We'll need to come up with definitions
that actually make sense.

Note that there is a temporary workaround for op-deployer image:

We need http locator support, which has been merged into op-deployer,
but the current version at HEAD doesn't quite work with kurtosis yet.

We'll fix that separately, at which point we'll be able to point to a
localDockerBuild of "op-deployer" just as well, if we want to.

* ct/scripts/ChainAssertions: add require revert message
Signed-off-by: jsvisa <delweng@gmail.com>

* ct/scripts/deploy/Deploy: add require revert message
Signed-off-by: jsvisa <delweng@gmail.com>

* ct/scripts/deploy/DeployOwnership: add require revert message
Signed-off-by: jsvisa <delweng@gmail.com>

* ct/scripts/deploy/DeployPeriphery: add require revert message
Signed-off-by: jsvisa <delweng@gmail.com>

* feat(semgrep): enable for all solidity files, except WETH9
Signed-off-by: jsvisa <delweng@gmail.com>

* fix require issues
Signed-off-by: jsvisa <delweng@gmail.com>

* fix(semgrep): run just semgrep to fix the require rules
Signed-off-by: jsvisa <delweng@gmail.com>

---------
Signed-off-by: jsvisa <delweng@gmail.com>

Co-authored-by: Tarun Khasnavis <tarunkhasnavis@Taruns-MacBook-Pro.local>

Added in extra dependency details to the getting started versions script to point people to where to learn more. (#13540)

* attempt...

* fixes

* fix checks

* fix kontrol build

* update DeploymentSummaryFaultProofs kontrol hash

* fixes

* undo unnecessary change

* feat: Reduce diff and usage of OptimismPortal2

* semver

* fix: init test

---------
Co-authored-by: Maurelian <maurelian@protonmail.ch>

* op-proposer: do not propose if outputRoot is the same as last proposed game

* fixed porposer tests

* PR comments

* Fix log message.

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

This handles deploying the kurtosis enclave according to spec, and
then stitching together the relevant information to be transmitted for
downstream consumption.

* feature: Allow adding arbitrary HTTP handlers to RPC HTTP Server.

* feature: Add sync package to handle streaming db files between nodes.

* tweak: Add DB sync server handler to op-supervisor server.

* feature: Use db sync if it's configured and there's no existing db.

* tweak: Use retry lib, which also adds max retries.

* rename: SyncRPC -> SyncEndpoint.

* tweak: Make the initial chain set part of the Server constructor.

* cleanup: Remove unused functions.

* rename: client -> syncClient to avoid package collision.

* fix: Load DepSet and use for sync server.

* tweak: Use client.BasicHTTPClient for sync client.

* fix: Don't double concat the path.

* tests,fix: Set sync resume to true.

* tweak: Add error context.

* tweak: Make http body close lint-friendly.

* tweak: Change retry policy to backoff from 1s to 30s, up to 10 minutes.

* tests,fix: Use require.ErrorIs to handle wrapped error.

This is in service of outputting useful information at the end of the
deployment.

Here we get back some useful wallets and addresses for downstream
consumption.

This enables us to retrieve information about artifacts and services,
for downstream consumption.

* Update README.md

* Update README.md

* op-supervisor: fix typo in README.md

---------
Co-authored-by: protolambda <proto@protolambda.com>