feat: LSP hover for integer literals

[asterite]

Description

Problem

No existing issue, just something Rust Analyzer does that I thought would be nice to have.

Summary

(it's better to review this commit by commit as the first one just moves things around)

Now hovering over an integer literal will show its inferred type together with its decimal value and hexadecimal value (Rust Analyzer also shows the binary (0b...) literal but I omitted that because Noir doesn't have such literals).

Additional Context

This PR also makes it easier for later PRs to provide hover over types and other AST nodes, to for example explain what Expr is, what i32 is, etc. (also similar to Rust Analyzer, which I think is nice because you can learn a lot of the language through the editor hover).

Documentation

Check one:

• No documentation needed.
• Documentation included in this PR.
• [For Experimental Features] Documentation to be submitted in a separate PR.

PR Checklist

• I have tested the changes locally.
• I have formatted the changes with Prettier and/or cargo fmt on default settings.

[michaeljklein]

Please add a few sanity-check tests for this since we could change a representation and end up showing the wrong hex/decimal value (I'm thinking of the upcoming #7324 changes).

Otherwise LGTM

[asterite]

I think when we change to SignedField we'll have to unpack that to pass it to the visitor, or make the visitor use SignedField. I'm not sure it could break without first stopping to compile 🤔

What would the test be like?

[TomAFrench]

LGTM, one nit on odd-length hexadecimal.

Agree that if we're changing how we're representing signed integers then this will break rather than failing silently. Maybe we could add another case within hover_on_integer_literal which points at a negative value.

Beyond that, I think the most we'd want to test for this is a couple of sanity checks that we get the expected formatted string given a (FieldElement, bool) input (to catch stuff like odd-length hex strings, etc.)

[asterite]

I think the most we'd want to test for this is a couple of sanity checks that we get the expected formatted string given a (FieldElement, bool)

Any numbers in particular that we should test? The logic mostly relies on format strings.

[TomAFrench]

🤷 Not really... something positive, something negative and zero.

[TomAFrench]

LGTM, the failures are unrelated to this PR. @michaeljklein can you remove the block on this?

[github-actions]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Test Suite Duration'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.

Benchmark suite	Current: 18e17e7	Previous: 582f56e	Ratio
AztecProtocol_aztec-packages_noir-projects_noir-protocol-circuits_crates_blob	67 s	51 s	1.31

This comment was automatically generated by workflow using github-action-benchmark.

CC: @TomAFrench