[group key addrs 7/7]: send and receive support for V2 addresses

[guggero]

This PR introduces a new V2 address that:

• Allows grouped assets to be sent.
• Allows the amount to be defined as zero by the recipient, allowing the sender to choose the amount.
• Can be re-used indefinitely without any privacy downsides.
• Use a new proof courier type to transfer an encrypted send fragment from the sender to the receiver using a mailbox system (but a more robust mailbox than the previous hashmailbox implementation).

---

This change is 

[guggero]

Okay, the itest failures are constant now, not flakes. Must be something I changed in the last push. Will address first thing tomorrow.

[guggero]

Found the itest issue, should be good to merge now.

[Roasbeef]

LGTM 🦜

Reviewed 43 of 63 files at r10, 20 of 20 files at r11, all commit messages.
Reviewable status: all files reviewed, 60 unresolved discussions (waiting on @ffranr, @guggero, and @ZZiigguurraatt)

---

itest/addrs_v2_test.go line 50 at r11 (raw file):

		amountToRequest := uint64(0)
		amountToSend := a.Amount - 1
		if amountToSend == 0 {

What type of error would get user get if they attempted to send a value greater than 1 to a collectible address? I wonder if we should encode that extra bit into the addr type.

FWIW, calls to decode the address should fetch the genesis as needed allowing this full information to be displayed.

---

tapdb/sqlc/queries/authmailbox.sql line 52 at r6 (raw file):

    m.encrypted_payload,
    m.arrival_timestamp,
    m.expiry_block_height,

Ah so we're removing this everywhere after all in favor the the should-be-good-enough-for-everyone value? np, we can always re add it later in a migration if needed

---

rpcserver.go line 1689 at r11 (raw file):

	// If the user didn't specify a specific address version but is using
	// a group key, we default to version 2.

👍

---

proof/courier.go line 1878 at r11 (raw file):

		// RPC methods that we require.
		statusErr, ok := status.FromError(err)
		if ok && statusErr.Code() == codes.Unimplemented {

Nice, I assume this is for cases where a Universe server hasn't updated to the latest version?

---

cmd/commands/addrs.go line 122 at r11 (raw file):

		// at least a V2 address. So we can just use that if the user
		// specifies no explicit version.
		if len(groupKey) > 0 {

👍

[guggero]

Reviewable status: all files reviewed, 60 unresolved discussions (waiting on @ffranr, @Roasbeef, and @ZZiigguurraatt)

---

itest/addrs_v2_test.go line 50 at r11 (raw file):

Previously, Roasbeef (Olaoluwa Osuntokun) wrote…

What type of error would get user get if they attempted to send a value greater than 1 to a collectible address? I wonder if we should encode that extra bit into the addr type.

FWIW, calls to decode the address should fetch the genesis as needed allowing this full information to be displayed.

Good question!
So due to this:

taproot-assets/address/address.go

Line 219 in 07d0c50

case asset.Collectible:

We don't allow addresses for collectibles to have a zero (optional) amount in the first place. So the error would already be thrown at address creation time: unable to make new addr: unable to make new addr: address: collectible asset amount not one.

I've added an integration test case for this!

---

proof/courier.go line 1878 at r11 (raw file):

Previously, Roasbeef (Olaoluwa Osuntokun) wrote…

Nice, I assume this is for cases where a Universe server hasn't updated to the latest version?

Yes, correct.

---

tapdb/sqlc/queries/authmailbox.sql line 52 at r6 (raw file):

Previously, Roasbeef (Olaoluwa Osuntokun) wrote…

Ah so we're removing this everywhere after all in favor the the should-be-good-enough-for-everyone value? np, we can always re add it later in a migration if needed

Yes, exactly. We can always add something later once we know exactly how/when we can do a cleanup without affecting emergency recovery.