FXPoolDeployer

[schystz]

Description

Introducing our new permissionless FXPoolDeployer, which will replace the older FXPoolFactory

Using the new FXPoolDeployer, we don't have to maintain a hardcoded list of oracle aggregator addresses anymore. Every time a pool is created, we simply fetch the aggregator address from FXPool.assimilator() -> Assimilator.oracle() -> Oracle.aggregator(). This isn't possible with the older FXPoolFactory because the pool needs to be initialized first before the assimilator is attached to the pool, which happens after NewFXPool event.

Other changes

• fixed oracle aggregator address for VCHF avalanche

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Dependency changes
• Code refactor / cleanup
• Documentation or wording changes
• Other

How should this be tested?

We've deployed the changes over to our balancer subgraph mainnet clone (trimmed down version):
https://thegraph.com/hosted-service/subgraph/xave-finance/balancer-v2

Simply run the "Query FXPools" from the query dropdown and it should return:

• DAI/USDC pool created by FXPoolFactory
• EURS/USDC pool created by FXPoolFactory
• GBPT/USDC pool created by FXPoolDeployer

Checklist:

• I have performed a self-review of my own code
• I have requested at least 1 review (If the PR is significant enough, use best judgement here)
• I have commented my code where relevant, particularly in hard-to-understand areas

dev -> master

• I have checked that all beta deployments have synced
• I have checked that the earliest block in the polygon pruned deployment is block, date/time
  • The earliest block is more than 24 hours old
• I have checked that core metrics are the same in the beta and production deployments

Merges to dev

• I have checked that the graft base is not a pruned deployment