[BUG] [v0.0.7] truncate_text() in summarizer.rs panics on multi-byte UTF-8 due to byte-index slicing #186

Bug Description

Location: src/cortex-compact/src/summarizer.rs:129

Type: Potential panic on multi-byte UTF-8 content

Version: v0.0.7

Issue Details

The truncate_text function uses byte-index slicing (&text[..max_chars]) which can panic if max_chars falls in the middle of a multi-byte UTF-8 character.

fn truncate_text(text: &str, max_tokens: usize) -> String {
    let max_chars = max_tokens * 4;
    if text.len() <= max_chars {
        text.to_string()
    } else {
        format!("{}...[truncated]", &text[..max_chars])  // BUG: byte-index slicing
    }
}

Why This Is a Bug

When text contains multi-byte UTF-8 characters (like emojis, Chinese characters, etc.), using byte-index slicing &text[..max_chars] can panic at runtime if max_chars lands in the middle of a multi-byte character.

Expected Behavior

The function should use character-based truncation (similar to what's done in cortex-common/src/truncate.rs) instead of byte-index slicing.

Suggested Fix

Use the safe truncation functions from cortex_common::truncate module:

use cortex_common::truncate::truncate_with_ellipsis;

fn truncate_text(text: &str, max_tokens: usize) -> String {
    let max_chars = max_tokens * 4;
    truncate_with_ellipsis(text, max_chars).into_owned()
}

Or implement character-based truncation inline using .chars().take().