docs(changelog): record post-PR60 frontend wiring trio (#64, #65, #66)

[ohgeeceee]

Doc-only patch: the [Unreleased] section of CHANGELOG.md was missing the merged trio (#64, #65, #66) that wires the u8_enum story end-to-end. Backend #60 already had a bullet; the gauge rendering, test harness, and unknown-byte sentinel merged without entries.

Just two short bullets, one for #65 and one for #66. PR #64 was already documented.

This is the smallest possible PR. No code touched. CI shouldn't change. Auto-merge should be safe per CLAUDE.md (doc-only).

[github-actions]

🤖 Doc-only change — auto-merge enabled. Will merge once required checks pass.

[chatgpt-codex-connector]

💡 Codex Review

beemuu/package.json

Line 12 in 7267097

"test:histogram": "node --test src/js/*.test.js"

Restore valid JSON in package manifest

This new script entry is missing a comma before the next property, so package.json is no longer parseable JSON. Any npm/pnpm command that reads the manifest (including running the new tests or building the Tauri app from the JS side) fails before scripts can execute.

---

beemuu/ops/backup-beemuu-db.sh

Lines 39 to 40 in 7267097

	elif [[ -n "${BEEMUU_DATA_DIR:-}" ]]; then
	DB_PATH="${BEEMUU_DATA_DIR%/}/beemuu.db"

Stop overwriting the resolved backup DB path

When production relies on BEEMUU_DATA_DIR=/var/www/beemuu/data as documented, this branch computes the correct live DB path, but the old DB_PATH="${BEEMUU_DB_PATH:-/var/www/beemuu/backend/data/beemuu.db}" assignment below immediately overwrites it unless BEEMUU_DB_PATH is also set. The cron will therefore fail to find the DB, or back up a stale in-repo DB, on the new prod layout this change is meant to support.

---

beemuu/src/js/main.js

Line 1621 in 7267097

row += "," + window.LiveFormat.csvCell(s.getAllData()[i]);

Avoid writing each log sample twice

When exporting a log CSV, the code above already appends one cell for this series; this additional append writes a second cell for the same sample while the header still contains only one column per series. With any enabled channel, every data row has too many fields and downstream CSV import shifts columns, so keep either the manual formatting or the shared LiveFormat.csvCell path, not both.

---

beemuu/src-tauri/src/data/live.rs

Line 341 in 7267097

Decode::U8Enum => "u8_enum",

Preserve enum maps when exporting profiles

Once profiles can serialize decode = "u8_enum", exporting a B58/N55 community profile through export_profile loses the parameter's enum_map: profile_to_toml still writes only id/label/unit/target/query/decode/min/max. Re-importing that exported TOML leaves the enum map empty, so live gauges show only 0xNN ? instead of the gear/engine-state labels.

---

beemuu/src/js/main.js

Line 777 in 7267097

? `${sf.label} (${r.moduleLabel})`

Read the serialized snake_case module label

For any service function with more than one routine, Rust serializes ModuleRoutine.module_label as module_label (the rest of this frontend reads returned Rust fields like fault_count and mileage_km in snake_case), so r.moduleLabel is undefined. The multi-module rows added here will display and confirm actions as Service (undefined) instead of showing the target module.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".