feat: allow optional precalculated stats to be used with gini, kurtosis and atkinson index

[Copilot]

Adds optional precalculated statistic parameters to gini, kurtosis, and atkinson functions to avoid redundant computation when these values are already known from prior calculations.

Changes

• gini(it, precalc_sum) - accepts optional pre-computed sum
• kurtosis(it, precalc_mean, precalc_variance) - accepts optional pre-computed mean and/or variance
• atkinson(it, epsilon, precalc_mean, precalc_geometric_sum) - accepts optional pre-computed mean and/or geometric sum (used when ε=1)

When precalculated values are None, functions compute them as before. When provided, they skip the corresponding computation pass.

Usage

use stats::{kurtosis, atkinson, gini};

let data = vec![1, 2, 3, 4, 5];

// If you already have mean and variance from a previous computation
let mean = 3.0;
let variance = 2.0;
let kurt = kurtosis(data.iter(), Some(mean), Some(variance));

// Or compute everything from scratch
let kurt = kurtosis(data.iter(), None, None);

// Similarly for atkinson with epsilon=1 when you have geometric_sum
let geometric_sum = data.iter().map(|&x| (x as f64).ln()).sum();
let atk = atkinson(data.iter(), 1.0, Some(mean), Some(geometric_sum));

// And for gini when you have the sum
let sum = 15.0;
let g = gini(data.iter(), Some(sum));

Tests verify that precalculated values produce identical results to computed values.

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.