cardano-testnet | Fix flaky height check in rpc test

[carbolymer]

Description

The chain height check was relying on an assumption that transaction goes into the next block, and sometimes the block nr check was returning stale info.

This changes the implementation to use retry untilM, to reliably check for the height.

Checklist

• Commit sequence broadly makes sense and commits have useful messages
• New tests are added if needed and existing tests are updated. These may include:
  • golden tests
  • property tests
  • roundtrip tests
  • integration tests
    See Running tests for more details
• Any changes are noted in the CHANGELOG.md for affected package
• The version bounds in .cabal files are updated
• CI passes. See note on CI. The following CI checks are required:
  • Code is linted with hlint. See .github/workflows/check-hlint.yml to get the hlint version
  • Code is formatted with stylish-haskell. See .github/workflows/stylish-haskell.yml to get the stylish-haskell version
  • Code builds on Linux, MacOS and Windows for ghc-9.6 and ghc-9.12
• Self-reviewed the diff

Note on CI

If your PR is from a fork, the necessary CI jobs won't trigger automatically for security reasons.
You will need to get someone with write privileges. Please contact IOG node developers to do this
for you.

[Copilot]

Pull request overview

This PR addresses flakiness in the Cardano testnet RPC transaction integration test by replacing a brittle “wait for next block height” approach with a retry-based check that waits until the expected UTxOs are observable after submission.

Changes:

• Replace the previous block-height polling loop with retryUntilM to wait until the destination address shows the expected UTxO state.
• Introduce an EpochStateView in the test to drive block-based retry timing.
• Add a changelog fragment documenting the fix.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
cardano-testnet/test/cardano-testnet-test/Cardano/Testnet/Test/Rpc/Transaction.hs	Reworks the post-submit waiting logic to retry querying UTxOs until the expected results appear, avoiding stale height assumptions.
cardano-testnet/changelog.d/20260424_081135_mgalazyn_fix_rpc_transaction_test.md	Documents the flakiness fix in the changelog fragments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

The note text says this checks that the submitted tx id is "in the submitted transactions list", but the assertion only compares the locally computed tx id to the id returned by submitTx. Consider updating the message to reflect what’s actually being verified (or add an RPC query to check presence in a list if that’s the intent).

Suggested change

	H.note_ "Ensure that submitted transaction ID is in the submitted transactions list"
	H.note_ "Ensure that submitTx returns the same transaction ID as the locally computed signed transaction ID"

[palas]

Nice 👍

[palas]

This assumes the testnet generated address has a single UTxO, and it seems that is indeed the case. But I think it would be more robust to either use a freshly generated address or check the assumption. Also the last assertion ends up being elem instead of a ===.

Also, now I realise this is really not from this PR, it was in the code already