Incorrect event state due to to indexing issues on Etherscan/Blockscout backends

[rkalis]

Today I noticed that BscScan especially has issues with returning logs for recent blocks. Combined with our local indexing of events, this can be problematic since there is no way to recover once BscScan gets back up to speed, because we have no indication that it "failed" earlier. I've reached out to the BscScan team to see what can be done about this. If that is not fruitful we just have to apply the same blockNumber - 50 calculation that we do for Covalent.

Since we don't know when this has occurred and who this affected we will need to add a Dexie migration for all Etherscan/Blockscan chains to clear all events at least since IndexedDB indexing went live, potentially just do a full clear.

[rkalis]

Also, certain unreliable RPCs for smaller chains may some times run into issues. We should try to figure out if there's a way to detect this kind of behaviour and make sure it doesn't break things - or can recover if things do break.