[12.x] Add percentile method to Collection

[nathanwuiske]

Summary

This PR introduces a percentile() helper to Illuminate\Support\Collection, filling the gap between the existing median() and mode() statistics helpers.

By computing percentiles, we can quickly understand the distribution of a dataset - whether it’s latency measurements, order totals, or any numeric series—and make decisions based on relative standing.

---

Why?

• Dashboards, alerts, and reports frequently rely on P95 latency, P90 response time, or top-10% spenders - all straightforward percentile queries.
• Collections already ship with mean, median, and mode. Percentiles are the logical missing metric in the basic descriptive-statistics toolkit.

---

Usage

// 1) P90 of students’ average grades:
$students = collect([
    ['name' => 'Alice', 'grades' => [88, 92, 85]],
    ['name' => 'Bob',   'grades' => [75, 78, 82]],
    ['name' => 'Cara',  'grades' => [95, 98, 100]],
]);

$p90Avg = $students
    ->map(fn($s) => collect($s['grades'])->avg())
    ->percentile(0.9);  // ≈ 95.9

// 2) Top-20% order totals:
$orderTotals = collect([45.00, 79.99, 120.00, 55.50, 65.00]);
$threshold   = $orderTotals->percentile(0.8); // e.g. 75.00

[github-actions]

Thanks for submitting a PR!

Note that draft PR's are not reviewed. If you would like a review, please mark your pull request as ready for review in the GitHub user interface.

Pull requests that are abandoned in draft may be closed due to inactivity.