fix(proof_registry): declare Axelar gateway and authorized-source keys (#379 #381 #382 #383)

[chideraisiguzor]

proof_registry: declare the Axelar gateway and authorized-source keys (#381)

This PR delivers one acceptance-criteria item from #381. #382 and #383 are referenced so that they close with this PR, but nothing from them is implemented here.

#381 Authenticate receive_message_axelar through the Axelar Gateway

What existed: initialize wrote ProofKey::AxelarGateway, and receive_message_axelar called Self::get_authorized_axelar_source(..), but neither the key variant nor the helper existed on main. They were lost in merge debris, so these were 2 of the crate's compile errors. There was also no way for the admin to configure an Axelar source.

Done:

• ProofKey::AxelarGateway and ProofKey::AuthorizedAxelarSource(Symbol) declared.
• Admin setters set_authorized_axelar_source(chain_name, source_address) and remove_authorized_axelar_source(chain_name). Both are admin-auth gated, bump the instance TTL, and emit axelar_source_authorized / axelar_source_removed.
• Getters get_authorized_axelar_source(chain_name) (the missing helper, now a contract entry point) and get_axelar_gateway().
• Test fixture now passes a gateway to initialize, which takes 3 args on main. The existing tests still called it with 2.
• 9 new tests: gateway recorded at init; source set/get/unset/per-chain/remove; setters rejected without admin auth; receive_message_axelar accepts the configured source and rejects a wrong source and an unconfigured or removed chain.

Not done in this PR:

• command_id + gateway.validate_message(..) through a contractclient trait (the forged-call exploit is still open).
• Fail-closed behaviour when the gateway is unset, and a timelocked gateway setter.
• SECURITY.md advisory and CHANGELOG entry.

#382 Replay protection and chain-identity binding on the Axelar path

Not done in this PR:

• SeenAxelar(command_id) replay key and its TTL.
• Axelar chain name to Wormhole chain id mapping with EmitterChainMismatch.
• Replacing the raw payload.get(i) calls with the checked byte() helper.
• command_id in ProofRecord, and the shared store_proof helper.
• docs/124 update.

#383 Restore the proof_registry pause circuit breaker

Not done in this PR:

• pause / unpause with admin + guardian and per-bridge ProofKey::Paused(bridge).
• Pause checks in both receive paths.
• Exported-spec regression test, pause matrix test, CHANGELOG, and the git-archaeology note. (The pause from 6cd974d was dropped by the merges f88a51b / 0ea03d3 / 6a3814c.)

#379 Per-token outflow rate limits as an automatic circuit breaker

Assigned after this PR was opened; folded in here (one PR per account + repo).

Not done in this PR:

• Per-token (window_secs, max_outflow) config behind a timelock (loosening timelocked, tightening immediate).
• Fixed-bucket or leaky-bucket outflow accumulator on contract-held funds (solver-to-user fills stay unlimited).
• Guardian / admin lift of the breaker.

Verification

proof_registry does not compile on main (pre-existing merge debris, unrelated to this change). It is missing PROOF_TTL_* / INSTANCE_TTL_* constants and the ChainIdOutOfRange / ProofStale variants, receive_message_axelar casts Option<u8> with as, and the #[cfg(feature = "testutils")] mock fns break #[contractimpl]. Because intent_settlement depends on it, that crate fails too.

To test this change I applied a minimal repair of that debris locally (not part of this PR) and ran it with this commit on top:

• cargo test --lib in proof_registry: 28 passed, 0 failed (19 existing + 9 new).
• cargo fmt --check and cargo clippy --all-targets: no findings on lines this PR touches. The remaining fmt hunks and 3 clippy warnings are on pre-existing lines.
• Doctests: 6 pre-existing failures from the untagged ``` block in receive_message_axelar's doc comment (not touched).

Closes #379
Closes #381
Closes #382
Closes #383

[drips-wave]

@chideraisiguzor Great news! 🎉 Based on an automated assessment of this PR, the linked Wave issue(s) no longer count against your application limits.

You can now already apply to more issues while waiting for a review of this PR. Keep up the great work! 🚀

Learn more about application limits