fix: invariant tests

[hexshire]

🤖 Linear

Closes OPT-183

[0xDiscotech]

• Missing invariant test of property 3. Is that on purpose or we're missing it?

• On fuzz_BurnLockedUSDC, isn't missing a check on the catch? currently we're doing assert(false) but shouldn't we check that the amount is not burned?
  Also, what about checking that it can't be burned if the state is not deprecated? Maybe I'm wrong bc I'm not used to echidna, but aren't we testing only the happy path there?

• A similar behaviour to the point mentioned above is present on fuzz_l1LinkedAdapterIncommingMessages and fuzz_l2LinkedAdapterIncommingMessages, where the catch clause is empty. Shouldn't we be testing that the message never suceeded andbecause of that the state didn't change?

• A similar empty catch scenario exists on fuzz_onlyMigrateToNativeForOwnershipTransfer

Can't comment because they werent modified:

• For consistency we should use 9_000_000 instead of 3_000_000 as gas limit when executing the deployments

Here is ensure:

Also check sometimes you used Insura and Insure again

Here remove the 9

[linear]

OPT-183 Modify current setup to make it compile and run again.

[hexshire]

@0xDiscotech > * Missing invariant test of property 3. Is that on purpose or we're missing it?

"Assuming the adapter is the only minter the amount locked in L1 should always equal the amount minted on L2"

This is something that should be modify since now the total might be locked on l1 == totalSupply + blacklistedFunds

[drgorillamd]

nit> don't leave an empty box for a new prop (so we know it still need to be at least tried;)