Replace PaymentAddress with MASP when transferring to a shielded address

[yito88]

Describe your changes

Checklist before merging

• If this PR has some consensus breaking changes, I added the corresponding breaking:: labels
  • This will require 2 reviewers to approve the changes

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 61.27%. Comparing base (8e04d45) to head (b82d487).
Report is 226 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #3626      +/-   ##
==========================================
- Coverage   61.27%   61.27%   -0.01%     
==========================================
  Files         312      312              
  Lines      101338   101338              
==========================================
- Hits        62092    62090       -2     
- Misses      39246    39248       +2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[cwgoes]

LGTM at least, @grarco should review as well

[yito88]

The test of an IBC shielding transfer failed when --disposable-gas-payer. I'm investigating it.

[yito88]

The test failed because the MASP VP rejected the transaction.
It seems that we can't replace it because the MASP VP checks the receiver address in the IBC packet for replay protection.

[grarco]

I've reverted the previous changes that were wrong because they were modifying the ibc receiver when Namada is the source chain when instead we want to do that only when we receive an ibc packet. At the moment though, we can't achieve this for Namada2Namada IBC shielded transfers cause the memo can be used only for either the shielding or the refund and in this case we'd need both. We could update build_ibc_transfer after we've changed the refunding to a transparent trasfer.

For now the only thing I did was updating the gaia e2e tests to use the MASP internal address instead of the payment address as the receiver in the IBC packet.

We should also update the docs somewhere to instruct users of other IBC chains to follow this practice when they submit a shielding ibc transaction to Namada (this is probably something for @brentstone)

@yito88

[grarco]

@cwgoes I ran the test locally with a wrong hardcoded address and the test failed. I can confirm that we convert the string that we get in the ibc packet into an Address using this function, which indeed casts all payment addresses to the internal MASP address:

namada/crates/core/src/address.rs

Lines 396 to 412 in 125af64

	impl TryFrom<Signer> for Address {
	type Error = DecodeError;
	fn try_from(signer: Signer) -> Result<Self> {
	// The given address should be an address or payment address. When
	// sending a token from a spending key, it has been already
	// replaced with the MASP address.
	Address::decode(signer.as_ref()).or(
	match crate::masp::PaymentAddress::from_str(signer.as_ref()) {
	Ok(_) => Ok(MASP),
	Err(_) => Err(DecodeError::InvalidInnerEncoding(format!(
	"Invalid address for IBC transfer: {signer}"
	))),
	},
	)
	}
	}