op-e2e: Nuke allocs

[mslipper]

This PR integrates op-deployer with op-e2e for the purposes of generating L1/L2 allocs on-the-fly. As a result, there is no longer any need to run make devnet-allocs or otherwise continue maintaining the legacy Python devnet. The generated allocs are at parity with those generated by the Python devnet - i.e., they contain the Alphabet VM, the fast fault game, and the deploy config customizations in devnetL1.json.

One notable change here is that the ability to generate new allocs files for L2OO chains will no longer be possible post-Holocene. To continue supporting the L2OO tests for the time being, the L2OO allocs files have been archived at the Holocene fork.

Cleaning up the old devnet will be handled separately.

[codecov]

Codecov Report

Attention: Patch coverage is 73.22835% with 34 lines in your changes missing coverage. Please review.

Project coverage is 47.34%. Comparing base (3c27ab2) to head (353275e).
Report is 18 commits behind head on develop.

Files with missing lines	Patch %	Lines
op-chain-ops/foundry/allocs.go	0.00%	8 Missing ⚠️
op-deployer/pkg/deployer/artifacts/locator.go	0.00%	6 Missing ⚠️
op-deployer/pkg/deployer/pipeline/dispute_games.go	79.31%	4 Missing and 2 partials ⚠️
op-deployer/pkg/deployer/opcm/anchor_roots.go	84.21%	2 Missing and 1 partial ⚠️
op-deployer/pkg/deployer/pipeline/opchain.go	84.21%	2 Missing and 1 partial ⚠️
op-service/util.go	0.00%	3 Missing ⚠️
op-deployer/pkg/deployer/opcm/l2genesis.go	0.00%	2 Missing ⚠️
op-deployer/pkg/deployer/pipeline/init.go	33.33%	0 Missing and 2 partials ⚠️
op-deployer/pkg/deployer/state/deploy_config.go	90.00%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop   #13364      +/-   ##
===========================================
+ Coverage    47.13%   47.34%   +0.20%     
===========================================
  Files          938      939       +1     
  Lines        78365    78464      +99     
  Branches       796      796              
===========================================
+ Hits         36940    37151     +211     
+ Misses       38697    38536     -161     
- Partials      2728     2777      +49     

Flag	Coverage Δ
cannon-go-tests-32	62.23% <ø> (ø)
cannon-go-tests-64	57.20% <ø> (ø)
contracts-bedrock-tests	91.76% <ø> (+2.67%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
op-chain-ops/interopgen/deploy.go	72.32% <100.00%> (+0.10%)	⬆️
op-deployer/pkg/deployer/inspect/l1.go	77.77% <100.00%> (+77.77%)	⬆️
...deployer/pkg/deployer/opcm/dispute_game_factory.go	100.00% <ø> (ø)
op-deployer/pkg/deployer/opcm/mips.go	100.00% <ø> (+33.33%)	⬆️
op-deployer/pkg/deployer/opcm/opchain.go	0.00% <ø> (ø)
op-deployer/pkg/deployer/state/chain_intent.go	82.35% <ø> (ø)
op-deployer/pkg/deployer/state/intent.go	45.37% <ø> (ø)
op-deployer/pkg/deployer/state/deploy_config.go	82.75% <90.00%> (ø)
op-deployer/pkg/deployer/opcm/l2genesis.go	0.00% <0.00%> (ø)
op-deployer/pkg/deployer/pipeline/init.go	59.40% <33.33%> (ø)
... and 6 more

... and 32 files with indirect coverage changes

[sebastianst]

Mainly reviewed the op-e2e init and setup code, which lgtm. A couple non-blocking comments and suggestions.

An alternative pattern to creating the allocs in init() could be to use lazy-initialization for each alloc type, so that they are created on demand the first time they are used, e.g. using sync.Once or sync.OnceValue.

[tynes]

This is a great step forward! Would be amazing if in the future we can decouple from the idea of allocs type and instead just run op-deployer in an EVM backend that can be cloned and handed off to different tests. This would let us skip needing to generate all of the allocs and write them to disk which would reduce a ton of boilerplate with setting up a hardfork, since there is 1 alloc file written to disk per hardfork