[mrorigo]

Problem

okf search panicked on documents containing non-ASCII text:

thread 'main' panicked at src/storage.rs:1056:20:
end byte index 330 is not a char boundary; it is inside 'ℂ' (bytes 328..331 of string)

snippet_from_doc computed byte offsets (pos - 40, pos + len + 80) from a lowercase match on the document body and sliced body directly. Those offsets can land inside a multi-byte UTF-8 character, and Rust panics when slicing at a non-char-boundary index.

Fix

• Added floor_char_boundary / ceil_char_boundary helpers that snap an arbitrary byte index to the nearest UTF-8 character boundary.
• snippet_from_doc now snaps both start and end, and guards against a reversed range.
• Added regression tests covering multibyte bodies for the direct match, offset-shifted match, and no-match fallback paths.

Verification

cargo fmt, cargo clippy --all-targets, and cargo test all pass.