Add key sub-command to 'derive' xpub

[notmandatory]

Description

This PR updates dependencies bdk to 0.4.0 and bdk-macros to 0.3.0 and adds a new key derive subcommand. The key derive sub-command derives an xpub from a given master xprv and derivation path.

The purpose of the derive command is to make it easier to create xpubs for use in testing multisig descriptors. For example an xpub can be derived and stored in an environment variable like this:

echo \"$ALICE_XPRV\"
"tprv8ZgxMBicQKsPfQjJy8ge2cvBfDjLxJSkvNLVQiw7BQ5gTjKadG2rrcQB5zjcdaaUTz5EDNJaS77q4DzjqjogQBfMsaXFFNP3UqoBnwt2kyT"
export ALICE_XPUB=$(bdk-cli key derive --xprv $ALICE_XPRV --path "m/84'/1'/0'/0" | jq -r ".xpub" | sed 's/^\[.*\]//g')
echo \"$ALICE_XPUB\"
"tpubDDrcq8JNTLVwaGnXp7KqZZxFJoic49ikEotWnpm6hpMY2hL2F1hpgbBBkEdsFJHU1iAx9DzMUKYr5mrphBMoXxhsBPhnVdmaoSCaeh6QWgj/0/*"

Checklists

All Submissions:

• I've signed all my commits
• I followed the contribution guidelines
• I ran cargo fmt and cargo clippy before committing

New Features:

• I've added tests for the new feature
• I've added docs for the new feature
• I've updated CHANGELOG.md

[thunderbiscuit]

A few questions on this. Does the derive subcommand always create an xpub at the last node in the provided --path?

If that is the case (say you give the path to be --path "m/84h/1h/0h/", would it then return an xpub that then needs to be extended (to add the account 0 and the *? In this case it ends up being 2 steps no matter what no?

I'm looking at the test and I'm not sure of how it knows which tpub to derive. Is the tpub the one associated with the account 0 or the one associated with the receive group of addresses (also 0 but unhardened)?

[notmandatory]

Yes, derive right now only creates an extended pub key and it always creates to the last hardened part of the path.. so if you use --path "m/84'/1'/0'/123" you'll get an tpub..../123/* , but if you use --path "m/84'/1'/0'" then the result is tpub..../*. So it's just deriving the minimum that's needed, since hardened parts of the path can only be derived from the extended private key.

[thunderbiscuit]

Ok I see! xpub happens at the last hardened node of the path, plus whatever extra derivation you want added on top. Thanks!

[thunderbiscuit]

I'm wondering if removing the default BIP32 root node xpub returned by the key generate subcommand should be part of this PR as well. The two feel related since this derive is a more flexible and better option for xpubs, and keeping the default one there in the generate subcommand only leads to more confusion IMO.

[notmandatory]

oh ya I should have mention it but I also removed the root xpub from the output of the generate and restore commands. :-)