feat(store): ERC-20 token registry

[Emmyt24]

Depends on: #214, #215. Blocks: #221, #225.

Description

On Stellar, a wallet's acceptable assets are knowable on-chain: a trustline is an explicit opt-in,
which is what crates/api/src/routes/trustlines.rs
exposes. ERC-20 has no equivalent — any address can receive any token, unsolicited, including
tokens designed to look like real ones.

There is also an existing wart worth fixing here: the testnet USDC issuer is a hard-coded constant
duplicated in two crates, with a comment acknowledging the duplication — see
USDC_TESTNET_ISSUER in crates/ingest/src/lib.rs and
crates/api/src/routes/payment_links.rs. A registry
subsumes both.

Build a per-chain token registry that is the single authority on what Octo will credit.

Requirements and context

• Registry rows are keyed by CAIP-19 asset id (AD-1) and carry chain_id, contract address,
  symbol, decimals, and enabled.
• decimals must be verified on-chain at registration by calling decimals(), not taken from
  operator input. USDC is 6 and DAI is 18; a wrong value here misprices every deposit of that token
  by a factor of 10^12.
• Unregistered tokens are never credited. They may be recorded for visibility (quarantine),
  matching how unattributable Stellar deposits are handled today.
• Registration is an admin-only operation. A user-registerable registry reintroduces exactly the
  attack it exists to prevent.
• Document the known-bad token classes explicitly, since operators will ask: fee-on-transfer
  (received ≠ event value), rebasing (balance changes without a transfer), and tokens with an
  upgradeable proxy or a blacklist/pause function that can freeze Octo's own treasury.
• Security: an attacker deploying a contract with symbol() == "USDC" costs nothing. Symbol is
  a display string with no uniqueness guarantee — match on contract address only, never symbol.

Suggested execution

Branch: feat/erc20-token-registry

Implement changes

• Migration 00NN_token_registry.sql: a tokens table keyed by CAIP-19 with the fields above,
  UNIQUE (chain_id, contract_address) on the lowercase-normalised address, and rows seeded for
  USDC on the target chains plus the existing Stellar assets so the hard-coded constants can be
  deleted.
• Store methods: register_token (admin), list_tokens(chain_id), get_token(caip19),
  is_creditable(chain_id, contract_address).
• On registration, call decimals() / symbol() via feat(evm-rpc): Resilient JSON-RPC client for EVM chains #218 and reject a mismatch against the
  submitted values rather than silently trusting either side.
• Replace both USDC_TESTNET_ISSUER constants with registry lookups.
• Add read-only API endpoints so clients can discover supported assets per chain.

Test and commit

• Test that registration rejects a decimals mismatch against the on-chain value.
• Test that a disabled token is not creditable and that an unregistered token is quarantined rather
  than credited or dropped.
• Test address normalisation: registering 0xABC... and looking up 0xabc... resolves.
• Test that two different contracts both reporting symbol() == "USDC" are distinct registry
  entries, and only the registered one is creditable.
• Test that the Stellar asset path still resolves the same USDC issuer it did via the constant —
  proving the constant removal is behaviour-preserving.
• Document the registry and the known-bad token classes in docs/architecture.md.

Example commit message

feat(store): per-chain ERC-20 token registry

ERC-20 has no trustline equivalent, so any address can receive any
token; the registry becomes the single authority on what Octo credits.

decimals() is read on-chain at registration rather than trusted from
input — USDC is 6 and DAI is 18, and a wrong value misprices deposits
by 10^12. Matching is on contract address only, never symbol, which
any attacker can spoof for free.

Also removes the duplicated USDC_TESTNET_ISSUER constants in
octo-ingest and octo-api.

Refs #223

[Peculiarr7]

Hi! I came across this issue and would love to work on it. I have experience with similar tasks and I'm confident I can investigate, implement a clean solution, and thoroughly test it before submitting a PR. If the issue is still open, I would appreciate it if you could assign it to me. I’ll prioritize this immediately and ensure a reliable solution is delivered as soon as possible. Thank you!

[Peculiarr7]

Hi! I came across this issue and would love to work on it. I have experience with similar tasks and I'm confident I can investigate, implement a clean solution, and thoroughly test it before submitting a PR. If the issue is still open, I would appreciate it if you could assign it to me. I’ll prioritize this immediately and ensure a reliable solution is delivered as soon as possible. Thank you!

[bobo-24]

Hi! I’m a full-stack developer with a strong blockchain background, and I’m confident I can resolve this issue quickly and effectively. I’ll deliver a clean, reliable solution with immediate attention to detail, ensuring it’s completed as soon as possible. Please assign it to me—I’m ready to get started right away.