[Bug] [v0.0.7] Align::pad in output.rs uses byte indexing text[..width] causing panic on multi-byte UTF-8 input

[echoforge2200]

Summary

Align::pad in src/cortex-engine/src/output.rs uses byte-based slicing text[..width] to truncate text when it exceeds the column width. This panics at runtime when the truncation point falls in the middle of a multi-byte UTF-8 character.

Location

File: src/cortex-engine/src/output.rs, lines 214-216

pub fn pad(&self, text: &str, width: usize) -> String {
    if text.len() >= width {
        return text[..width].to_string();  // BUG: byte index, not char boundary
    }

Root Cause

text.len() returns the byte length, not the character count. For multi-byte UTF-8 strings (e.g., containing Chinese, Japanese, Arabic, emoji, or accented characters), text[..width] will attempt to slice at a byte offset that may not be a valid UTF-8 character boundary, causing a panic (byte index N is not a char boundary).

Reproduction

let text = "héllo"; // 'é' is 2 bytes (0xC3 0xA9)
// text.len() == 6 (bytes), but char count is 5
// If width == 5, text[..5] slices mid-character -> PANIC
let result = Align::Left.pad("héllo", 5);

User-Facing Impact

The Align::pad function is used by TableBuilder::format_row (line 327) which is called by TableBuilder::build (line 304). Any user who renders a table containing non-ASCII text in a cell that exceeds the column width will experience a runtime panic, crashing the application.

Fix

Use text.chars().count() for length comparison and text.char_indices().nth(width) to find the correct byte boundary for truncation.