ci: update Bun versions in workflows

[DarkGL]

Description

Updates the Bun version matrix in GitHub workflows to use 1.2.12, removing older versions for consistency and alignment. Also refines the available Bun versions in documentation code to reflect these changes.

Testing

CI

Checklist

• Conducted a self-review of the code changes.

[Copilot]

Pull Request Overview

This PR updates the Bun version configuration across the project for consistency. Key changes include:

• Updating the Bun version in the documentation (docs/src/App.tsx) to use [1.2] (representing the major.minor version).
• Updating the Bun version matrices in both .github/workflows/release.yml and .github/workflows/pr.yml to use "1.2.12".

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File	Description
docs/src/App.tsx	Updated Bun version array to [1.2] to reflect the new supported version.
.github/workflows/release.yml	Changed the Bun version matrix from older versions to "1.2.12".
.github/workflows/pr.yml	Changed the Bun version matrix from older versions to "1.2.12".

Comments suppressed due to low confidence (1)

.github/workflows/release.yml:95

• [nitpick] Verify that the updated Bun version '1.2.12' in the workflow matrix consistently reflects the supported version documented in the code; a brief inline note could improve clarity for future maintenance.

            - 1.2.12

[DarkGL]

@hoeck @moltar

[hoeck]

Thanks for taking care of updating this.

But shouldn't the benchmark then only show results for the (only) bun version that is tested? Atm. it will show the (old) results for bun 1.1 too but they will never get updated if libraries are updated?

[DarkGL]

@hoeck

In this issue, #1860 @moltar mentioned can we keep historical data thats why I left it, what do you think ? Should we remove or leave it ? Personally, I'm on side for removing it for older bun versions

[hoeck]

@hoeck

In this issue, #1860 @moltar mentioned can we keep historical data thats why I left it, what do you think ? Should we remove or leave it ? Personally, I'm on side for removing it for older bun versions

Yes please remove it, as the current benchmark viewer should only show data for the most recent library + runtime combination to not confuse users.

The historical data is saved in the git repository via commits of all the different result json files. Someone could dig into that data and build a historical benchmark viewer later.

[DarkGL]

Done