Detect content types when browsing files

[andrew]

Use magic as the content-type fallback for browsed files without a known path mapping. The handler peeks at most 512 bytes and replays them while streaming the file, and only promotes PNG, JPEG, GIF, and PDF to registered MIME types. Active web formats remain text/plain.\n\nUpdate archives to v0.4.0 and remove the extensionless ZIP filename workaround so archives are detected from content.

[Copilot]

Pull request overview

This PR improves the file-browsing handler by adding content-based MIME detection for files that don’t have a reliable extension/path mapping, while keeping potentially active web formats served as text/plain. It also updates the archive handling dependency so extensionless archives can be detected from content instead of relying on filename workarounds.

Changes:

• Add a 512-byte prefix peek and magic-based content-type fallback for unknown-path browsed files (promoting PNG/JPEG/GIF/PDF only).
• Update github.com/git-pkgs/archives to v0.4.0 and remove the extensionless-zip filename workaround in archive opening.
• Expand browse tests and add a micro-benchmark for content-type detection.

Reviewed changes

Copilot reviewed 4 out of 5 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
internal/server/browse.go	Adds prefix-peek + magic fallback for MIME detection and removes archive filename extension workaround.
internal/server/browse_test.go	Adds coverage for extensionless/binary sniffing behavior and updates archive behavior tests.
internal/server/browse_bench_test.go	Adds a benchmark for the new content-type detection paths.
go.mod	Updates archives to v0.4.0 and adds magic dependency.
go.sum	Updates checksums for the new/updated dependencies.

---

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