fix(index): add guard for short files in magic number check

[Fdawgs]

fileHandle.read() may return fewer bytes than requested for files smaller than 6 bytes. Without checking bytesRead, the comparison against the RTF magic number {\rtf1 could produce a false positive on truncated input.

Also switches to Buffer.allocUnsafe as all bytes are immediately overwritten by the read, so the filling of all zeroes in Buffer.alloc was wasted. Teeny tiny micro-optimisation that saves ~2.84µs per call. All adds up when converting millions of files though!

Checklist

• Commit message adheres to the Conventional commits style, following the @commitlint/config-conventional config

[Copilot]

Pull request overview

This PR updates the RTF “magic number” detection in UnRTF.convert() to safely handle inputs shorter than the expected header length, and switches the header buffer allocation to an unsafe variant for a minor performance win while preserving correctness.

Changes:

• Add a bytesRead guard so files shorter than RTF_MAGIC_NUMBER_LENGTH can’t be misclassified.
• Switch from Buffer.alloc to Buffer.allocUnsafe for the magic-number read buffer.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 2 out of 3 changed files in this pull request and generated no new comments.