miette integration for all errors in Core, Validator, and cedar-policy

[cdisselkoen]

Description of changes

Derives (or manually implements) miette::Diagnostic for all our error types. Tries to make good use of the help() field, but does not yet assign error codes or propagate source locations; those can be done in followup PRs (and have their own GitHub issues).

No breaking changes to the public API. Some error messages have changed, and in particular, in some cases, information that used to be in the Display for an error is now in other miette fields. I have updated the CHANGELOG with a note about this. I'm not sure where else we should document it.

To display a miette::Diagnostic in a pretty fashion, convert it to miette::Report using From/Into. Note that ? automatically converts errors, so it is sufficient to declare the error type of your function (or main()) as miette::Report. Note also that you need the fancy feature of miette for prettiest output; we enable this in the CLI crate but not in our library crates, as recommended by miette. ("You should only [enable the "fancy" feature] in your toplevel crate, as the fancy feature pulls in a number of dependencies that libraries and such might not want.")

Sample outputs from the CLI:

Issue #, if available

Resolves #182

Checklist for requesting a review

The change in this PR is (choose one, and delete the other options):

• A bug fix or other functionality change requiring a patch to cedar-policy.

I confirm that this PR (choose one, and delete the other options):

• Updates the "Unreleased" section of the CHANGELOG with a description of my change (required for major/minor version bumps).

I confirm that cedar-spec (choose one, and delete the other options):

• Does not require updates because my change does not impact the Cedar Dafny model or DRT infrastructure.

Disclaimer

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

[john-h-kastner-aws]

Can we make the error just invalid escape pattern and rely on #[related] to aggregate the errors?

[cdisselkoen]

We could. Or we could just show the first error message in the #[error], like we do for parse errors. In #326 we decided not to totally kneecap the Display for ParseErrors in case users are only looking at the Display. We could revisit that decision, or decide differently here. Do you have a preference between no error message / first error message / all error messages?

[john-h-kastner-aws]

It doesn't bother me too much. Using the {:?} debug formatting in error messages doesn't always work out well, but if it formats nicely in this case it's fine

[cdisselkoen]

Fixed this up some, including the error message in unescape::UnescapeError itself

[john-h-kastner-aws]

Suggested change

	#[diagnostic(help("any common types used in entity or context attributes need to be declared in `commonTypes`"))]
	#[diagnostic(help("any common types used in entity or context attributes need to be declared in `commonTypes`, and common types may not reference other common types"))]

We could do more to make the error in this case nice, but this at least helps

[cdisselkoen]

is this the error we get if we try to have common types reference other common types?

if so I can update the doc comment on the variant as well

[john-h-kastner-aws]

That'd be good

[cdisselkoen]

Updated