improve(OP): More generic OP chain adapter deployment

[pxrl]

This change permits an incremental step towards a totally generic chain adapter deployment for OP stack chains. It does this be restructuring the address mappings for OP stack chains, such that the L1 contract addresses can be resolved solely with the combination of HubPool and SpokPool chain IDs. It also updates the new OP_Adapter deployment script to support a runtime-defined SpokePool chain ID. This is clunkily supplied via the environment because hardhat deploy doesn't seem to permit arbitrary runtime arguments being supplied, but that aside, this should ease subsequent deployments.

[pxrl]

nb. This only touches constants and deployment scripts, so there should be no audit sensitivity. Contract addresses are being moved around though, so it's important that they're correct. I've checked myself but it'd be good to have a 2nd set of eyes on it.

[bmzig]

I checked also and the bridge addresses check out.

[bmzig]

Looks good. I'm guessing we are just keeping the deploy_[op network]_adapter scripts for the sake of bookkeeping at this point?

[pxrl]

Looks good. I'm guessing we are just keeping the deploy_[op network]_adapter scripts for the sake of bookkeeping at this point?

Kinda, yeah. I do think we could additionally consolidate:

• Optimism & Base (OP stack, native USDC via CCTP)
• Zora & Mode (OP stack, bridged USDC)

Blast is a total snowflake of course, and Redstone now seems to have an upgradable USDC deployment here, so in theory we could remove the Redstone Adapter in favour of the OP Adapter (even irrespective of whether we support USDC there).

However...I probably don't expect many (or any) new chain deployments to fit the Optimism/Base pattern or the Zora/Mode pattern, so I think this optimisation for the OP Stack + upgradable USDC combination gives us the most benefit for the least effort.