feat: add eth_chain_id entrypoint

[bitfalt]

Add eth_chain_id entrypoint to the KakarotCore contract. Add a new constant to compute 2**53. Draft PR since replacing of all references is still needed, also want to ask feedback regarding implementation (correct types) to @enitrat.

Pull Request type

Please check the type of change your PR introduces:

• Bugfix
• Feature
• Code style update (formatting, renaming)
• Refactoring (no functional changes, no API changes)
• Build-related changes
• Documentation content changes
• Other (please describe):

What is the current behavior?

Resolves: #923

What is the new behavior?

Add an entrypoint to the KakarotCore contract get the eth_chain_id from the transaction

• Add a new constant in utils/src/constants.cairo for 2**53
• Add eth_chain_id entrypoint

Does this introduce a breaking change?

• Yes
• No

---

This change is 

[bitfalt]

I've checked the codebase and found 3 places where we can call kakarot.eth_chain_id(). I have some questions regarding how to call the function in some places, and would appreciate any guidance or feedback towards my solution!

kakarot-ssj/crates/evm/src/backend/starknet_backend.cairo

Line 82 in eeac8bb

chain_id: get_tx_info().unbox().chain_id.try_into().unwrap(),

I have question regarding the code above, since it is doing chain_id: get_tx_info().unbox().chaind_id.try_into().unwrap() which is similar to what the implementation of the function I did.
However, I'd like to know if it would be viable to call eth_chain_id. If the refactor is necessary here, I'd like to know how to call the function, since I'm used to work with dispatchers, but I don't see any address and the only reference to KakarotCore I see is kakarot_state.

kakarot-ssj/crates/contracts/src/account_contract.cairo

Lines 243 to 259 in eeac8bb

	let chain_id: u64 = tx_info.chain_id.try_into().unwrap() % POW_2_32.try_into().unwrap();
	assert(signature.len() == 5, 'EOA: Invalid signature length');
	let signature = deserialize_signature(signature, chain_id)
	.expect('EOA: invalid signature');
	let mut encoded_tx_data = deserialize_bytes((*outside_execution.calls[0]).calldata)
	.expect('conversion to Span<u8> failed')
	.span();
	let unsigned_transaction = TransactionUnsignedTrait::decode_enveloped(
	ref encoded_tx_data
	)
	.expect('EOA: could not decode tx');
	let address = self.Account_evm_address.read();
	verify_eth_signature(unsigned_transaction.hash, signature, address);
	let kakarot = IKakarotCoreDispatcher { contract_address: self.ownable.owner() };

The code above is easy to refactor, my solution would be move the line 259 (let kakarot = IKakarotCoreDispatcher { contract_address: self.ownable.owner() };) to be above line 243 (let chain_id: u64 = tx_info.chain_id.try_into().unwrap() % POW_2_32.try_into().unwrap();). Then just call the implemented function to set the chain_id. In this way: let chain_id: u64 = kakarot.eth_chain_id().

kakarot-ssj/crates/evm/src/backend/validation.cairo

Lines 19 to 24 in eeac8bb

	let kakarot_chain_id: u64 = get_tx_info()
	.chain_id
	.try_into()
	.unwrap() % POW_2_32
	.try_into()
	.unwrap();

Regarding the code above, I have the same question as the first one I made. I'm thinking that we could call eth_chain_id function in the KakarotCore contract. However, I'm facing the same issue were this is no address, but in this case we have the parameter kakarot_state. However, I still don't really get how to call the function eth_chain_id implemented in KakarotCore without the dispatcher.

Any guidance toward the questions above is appreciated!

[enitrat]

regarding 1. yes you should call eth_chain_id

for 1. and 3. the unsafe_new_contract_state function gives you the ContractState object of the contract you call in on - in this case, KakarotCore::ContractState. This means that any functions defined with self: ContractState in the kakarot contract can be called through this object. Make sure you correctly import the IKakarotCore trait, in that defines functions callable on the KakarotCore::ContractState type.

example:

/// Populate an Environment with Starknet syscalls.
pub fn get_env(origin: EthAddress, gas_price: u128) -> Environment {
    let kakarot_state = KakarotCore::unsafe_new_contract_state();
    let kakarot_storage = kakarot_state.snapshot_deref().storage();
    // let chain_id = kakarot_state.eth_chain_id();
    let block_info = get_block_info().unbox();

    // tx.gas_price and env.gas_price have the same values here
    // - this is not always true in EVM transactions
    Environment {
        origin: origin,
        gas_price,
        chain_id: get_tx_info().unbox().chain_id.try_into().unwrap(),
        prevrandao: kakarot_storage.Kakarot_prev_randao.read(),
        block_number: block_info.block_number,
        block_gas_limit: constants::BLOCK_GAS_LIMIT,
        block_timestamp: block_info.block_timestamp,
        coinbase: kakarot_storage.Kakarot_coinbase.read(),
        base_fee: kakarot_storage.Kakarot_base_fee.read(),
        state: Default::default(),
    }
}

[enitrat]

@bitfalt can you move the entrypoint to the EthRPC trait to sync with changes made in #937 and rebase on main?

[bitfalt]

@enitrat sure! I'll modify the code accordingly.

[enitrat]

lgtm, just add a unit test in the same file :)

[bitfalt]

okayy, wouldn't it be better to have the test in crates/contracts/tests/test_kakarot_core.cairo along with all the other tests?

nvm, i just saw how you're doing the other tests, will do the same!

[enitrat]

yes, in the same module, under a #[cfg(test)] module, as we're unit testing

[enitrat]

@bitfalt there is a merge conflict; can you rebase?

[bitfalt]

Just rebased everything, I think I didn't mess up the rebase this time. I'm fairly new to doing rebase instead of merge, so it is a bit confusing to me haha. Let me know if there is anything else!