Don't use Ethereum paths for testing networks

[prusnak]

It has been brought to our attention that Metamask uses m/44'/60'/x' paths even for testnet version of the networks (e.g. Rinkeby) - trezor/trezor-firmware#1474

For testnet networks, m/44'/1'/x' path should be used as per BIP44

[Gudahtt]

I'm assuming that you meant to link SLIP-0044? SLIP-0044 appears to suggest using 1 for all testnets, but BIP44 describes 1 as just being for the Bitcoin testnet.

In any case, using the same set of accounts for all networks is fairly fundamental to how MetaMask works at this point. If we were to break this, it would be very disruptive for many users. The portability of accounts between Mainnet and testnets is relied upon for testing, for example.

I'll leave this open for consideration for now, as we may consider this in some future major update.

[benjamincburns]

related: #3131