* op-node: Cleanup channel definition

This moves some files around and then expands upon the Channel struct.
It is better built for adding frames + then reading data when it is
ready.

Some tests are still failing and I don't know why.

* ReadBatch in channel.go

* op-node: Fix channel.go + start of new interface

* fix lint

* op-node: Use simple channel decoding

* revert switch to reader

This was due to the difficulty in making the mocks work with the reader.

* fix lint

* fix bug around trying to use an empty chanel
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

* state-surgery: layer 2 genesis block generation

Generate the L2 or L1 genesis block given a hardhat and a deploy config.

* state-surgery: some comments

* state-surgery: delete dead comment

* op-node: Rename channel_frame.go -> fame.go

* op-node: New ParseFrames function + fuzzers

* op-node: Use new ParseFrames function

* fix test to new behavior

* small fixes

* fix lint

* small fixes

* fix nits

* fix nits

* fix logging format
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

This helps improve performance.

Adds per-commit dev builds that get uploaded to our GCR repository. This enables running hive on a per-PR basis.

The op-geth repo has already been created as a fork of reference-optimistic-geth.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

state-surgery: fix string slice

state-surgery: fix naming

The `initialize()` function in the L2CrossDomainMessenger
sets some blocked system addresses. It was using `address(this)`
but in reality the contract is a predeploy at a specific address.
It should not be deployed, the `initialize` function will never
run. For the unit tests, the contracts are placed at their correct
addresses. This makes the testing environment as close as possible
to what is experienced on a live network.

This PR hardcodes in the L2 cross domain messenger constant
from the predeploys library instead of using `address(this)`.
In practice this value is set using the chain ops tooling,
so this makes it more clear as to what the value should be.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Do not overwrite accounts that already exist when
creating accounts with the memory db. Also panic
if trying to set state in an account that doesn't
yet exist in the state.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Someplace in foundry it seems to not populate the
key value correctly in the compiler storage layout type.
Unclear why its not always populated, it is in some cases
but not all. It could be anywhere between solc, foundry and
the foundry hardhat plugin.

The information is still there, it just needs to be parsed
from a different field. This PR uses a regex to parse the value
to pull it out. It recursively calls itself and we can be sure
that it will not infinitely recurse because a non empty string
is passed in the second time.

With this PR, I am able to successfully build the L2 genesis
state.

bugfix

fix
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Fix some naming so that it reads better

For some reason `strings.TrimPrefix` is removing
an extra character sometimes. We want to remove the
`.json` extension and using it will sometimes also
remove the final character in the contract name.
This is problematic because when you call
`hh.GetDeployment`, with the name of the contract,
it will not be able to find the deployment because
the name of the deployment will not match the name
passed in by the user. This instead slices off the
`.json` from the string manually.

- Add support for `&common.Address`
- Take locks in initialization of `hardhat.Hardhat`

contracts-bedrock: migrate deploy config to json

Migrates the deploy config from typescript to json
so that it can be read into a go program.

Adds the ability to serialize strings in state.
It only supports strings less than 32 bytes long
which would require major refactoring and we do
not currently need that functionality.

Main test cases come from mainnet WETH contract
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Updates the SDK to load contract addresses from the deployments folder.
Also updates the Contracts package to export addresses from the
deployments folder. More robust to address changes.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Makes it possible to easily deploy via PK if not deploying via Ledger.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Refactors contract addresses and related constants into the
chain-constants.ts file. Much cleaner separation of logic.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

state-surgery: implement bytes32 handling

state-surgery: fix flakey tests

contracts-bedrock: make a library call internal

state-surgery: handle fs errors

Sometimes `require.Equal` flakes on the comparison
of `big.Int`.

https://github.com/ethereum-optimism/optimism/blob/b31d35b67755479645dd150e7cc8c6710f0b4a56/op-node/rollup/derive/fuzz_parsers_test.go#L42Co-authored-by: Joshua Gutow <jgutow@optimism.io>

Adds the ability to set bytes32 values in state.
There is a single bytes32 value in our L2 predeploys.
Adds test coverage that the value is set properly and
encoded properly. Both `common.Hash` and hex strings
are supported to be able to be encoded into a bytes32
suitable for storage.

Fixes a bug in the SDK that would cause withdrawals to revert for
certain custom bridges on Kovan.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Handle errors when walking the filesystem so that they do
not silently fail.

So we don't have to handle library linking
in the state surgery

I forgot to initialize the forceReset channel which means that the
RPC isn't able to signal to the state loop that it needs to reset.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Removes the ICrossChainMessenger interface. We don't have a strong
reason to maintain this interface and it introduces complexity to the
process of developing the SDK. With the interface, we are required to
make changes to multiple files when we simply want to change the
CrossChainMessenger class. We are also not maintaing multiple
implementations of the CrossChainMessenger that would make the interface
useful.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Deploys Drippie to the Optimism Goerli testnet. Have not set any
configurations up yet, still waiting on Gelato support. Not verified on
Etherscan since we don't have an Etherscan for OP Goerli yet. Addresses
are different than on other chains because of the update to Solidity
0.8.15.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

Updates Drippie config for mainnet. A new drip was created after the
previous drip was archived.
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>

* predeploys: fix address names

Updates the address names and also create a
mapping of addresses so that they can be iterated
over.

* op-bindings: export predeploys mapping

* fix test
Co-authored-by: Matthew Slipper <me@matthewslipper.com>

contracts-bedrock: update @foundry-rs/hardhat-forge

Fixes the build issues in CI