feat: add median to --output-json

[Joristdh]

Description

In #5398 the JSON reporter got reworked for the benchmark results. In this rework, the samples array was dropped to prevent huge output files. However, without the samples it's not possible anymore to manually calculate some extra metrics after the bench run. The median is simple enough to calculate when having acces to the samples and can provide some interesting insights (e.g. significant difference between mean and median could indicate outliers).

Please don't delete this checklist! Before submitting the PR, please make sure you do the following:

• It's really useful if your PR references an issue where it is discussed ahead of time. If the feature is substantial or introduces breaking changes without a discussion, PR might be closed.
• Ideally, include a test that fails without this PR but passes with it.
• Please, don't make changes to pnpm-lock.yaml unless you introduce a new test example.

Tests

• Run the tests with pnpm test:ci.

Documentation

• If you introduce new functionality, document it. You can run documentation with pnpm run docs command.

Changesets

• Changes in changelog are generated from PR name. Please, make sure that it explains your changes in an understandable manner. Please, prefix changeset messages with feat:, fix:, perf:, docs:, or chore:.

[Joristdh]

If the Vitest framework isn't the right place for new metrics, the change could also be migrated to the underlying Tinybench results.

[sheremet-va]

If the Vitest framework isn't the right place for new metrics, the change could also be migrated to the underlying Tinybench results.

I guess it makes sense to have it in Vitest because you can extrapolate it in tinybench from samples, although it might also be nice for tinybench users to have a quick access to median on the object, so I am fine either way

[Joristdh]

I think indeed that calculating the median is trivial when having access to all the samples, so this mainly adds value for Vitest users.

Maybe it'd be even better to just add a flag to preserve the original samples. Then Vitest users can deal with/use the results however they see fit.

In any case, that would be something for another PR. Is there something I just do to get this merged after the approval @sheremet-va?