Fix nonce synchronization issues in EVM deployments by adding slow flag to forge

[evgeniko]

EVM deployments were failing with nonce errors on certain networks (e.g. Mezo) due to forge sending transactions too quickly without waiting for confirmation between transactions.
Example error:
server returned an error response: error code -32000: invalid nonce; got 3, expected 2: invalid sequence: invalid sequence

Added --slow flag to forge script commands in both deployEvm and upgradeEvm functions. This flag forces forge to wait for transaction confirmation before sending the next one, ensuring proper nonce synchronization.

[kcsongor]

If this issue only comes up for cetain chain, we should only add it for those chains, in order to not unnecessarily slow down deployments for other chains. AFAIK the fast-path works reliably everywhere else, right?

[nik-suri]

@evgeniko adding on to what @kcsongor said -- could we solve this by exposing the --slow flag as an option so CLI users can apply it on a case-by-case basis whenever needed?

[kcsongor]

I agree a --slow flag could make sense but I would argue it will be worse UX than just automatically applying it to chains that we know need it, because the error message is quite cryptic and it breaks the flow.

[evgeniko]

Added helper function that add slow flag only if Mezo is used as a hardcoded value.