Test checkunparsable errors #35834

CheckUnparsable parses both a private-form descriptor and its corresponding public-form descriptor. The helper previously reused one error string for both parses.

The public parse therefore overwrote the private parse’s error. Although the helper verified that both parses failed, its exact error assertion only checked the error produced by the public parse. An incorrect private-form error could consequently go undetected.

Changes

• Store the private and public parse errors separately.
• Check the exact private error when a private-form descriptor is provided.
• Continue checking the exact public error.
• Add an overload that accepts different expected private and public errors.
• Preserve the existing three-argument interface for cases where both errors should match.
• Update the affected test cases with their existing private-form errors.

This also corrects a raw descriptor test whose private input used raw(...) while its public input mistakenly used sh(...). Both inputs now use raw(...), with Raw script is not hex as the expected error.

This PR only changes test code and does not affect production descriptor parsing.

<!--e57a25ab6845829454e8d69fc972939a-->

The following sections might be updated with supplementary metadata relevant to reviewers and maintainers.

<!--006a51241073e994b41acfe9ec718e94-->

Code Coverage & Benchmarks

For details see: https://corecheck.dev/bitcoin/bitcoin/pulls/35834.

<!--021abf342d371248e50ceaed478a90ca-->

Reviews

See the guideline and AI policy for information on the review process. A summary of reviews will appear here.

<!--174a7506f384e20aa4161008e828411d-->

Conflicts

Reviewers, this pull request conflicts with the following ones:

• #35445 (wallet, descriptor: Revert StringType::COMPAT for Miniscript expressions and drop the concept of a Descriptor ID that can be validated by achow101)

If you consider this pull request important, please also help to review the conflicting pull requests. Ideally, start with the one that should be merged first.

<!--5faf32d7da4f0f540f40219e4f7537a3-->

<!--85328a0da195eb286784d51f73fa0af9-->

🚧 At least one of the CI tasks failed. _{Task riscv32 bare metal, static libbitcoin_consensus: https://github.com/bitcoin/bitcoin/actions/runs/30452867012/job/92449751882} _{LLM reason (✨ experimental): CI failed because submodule clones from sourceware.org during 01_base_install.sh hit HTTP 429 rate limits (“unable to access … returned error: 429”), causing the Docker build to abort (exit code 2).}

<details><summary>Hints</summary>

Try to run the tests locally, according to the documentation. However, a CI failure may still happen due to a number of reasons, for example:

• Possibly due to a silent merge conflict (the changes in this pull request being incompatible with the current code in the target branch). If so, make sure to rebase on the latest commit of the target branch.

• A sanitizer issue, which can only be found by compiling with the sanitizer and running the affected test.

• An intermittent issue.

Leave a comment here, if you need help tracking down a confusing failure.

</details>