Fix dual-battery handling in upower queries

[stefanomainardi]

Summary

Fixes the dual-battery inconsistency reported by @Michael-Steshenko in #5780: the waybar percentage and the right-click notification disagree on multi-battery systems (e.g. ThinkPad T480 with internal + slice).

Root cause

Five scripts share the same anti-pattern:

upower -i $(upower -e | grep BAT)

upower -e lists every device path. When two batteries exist, grep BAT returns two lines, and upower -i is invoked with multiple positional args — it only honors the first. So the scripts report BAT0 only, while waybar's native battery module shows the correct unified value.

Fix

Switch every call to upower -i /org/freedesktop/UPower/devices/DisplayDevice. That's upower's composite virtual device aggregating all power sources — the same data waybar uses internally. Same fields are exposed (percentage, energy-rate, state, time-to-empty/full, energy-full), so no other code change is needed.

Scope

All five scripts in the affected callgraph are fixed in a single commit because they all share the same bug from the same root cause:

Script	Used by
bin/omarchy-battery-status	waybar right-click + SUPER+CTRL+ALT+B notification
bin/omarchy-battery-remaining-time	called by battery-status
bin/omarchy-battery-capacity	called by battery-status
bin/omarchy-battery-remaining	called by battery-monitor
bin/omarchy-battery-monitor	systemd timer, critical-low alert

The last two aren't strictly required to fix the symptom in the issue, but the same anti-pattern was silently breaking the critical-low alert as well — on dual-battery systems it could misfire (trigger when only BAT0 hits 10% while BAT1 is full) or never trigger (BAT0 stays charged thanks to BAT1 draining first). Reviewers — happy to split if you prefer one PR per concern.

Master backport

The same five scripts exist on master with identical content (verified: git diff origin/master:bin/omarchy-battery-{status,remaining,remaining-time,capacity,monitor} origin/dev:...). If you'd like me to open a parallel PR against master, just say so.

Test plan

• bash -n clean on all five scripts
• Smoke-run on a single-battery (well, no-battery — DisplayDevice still exists and returns sensible empty values) system, scripts don't crash and produce parseable output
• Multi-battery verification — I don't own a ThinkPad T480-class machine; relying on @Michael-Steshenko's root-cause analysis (their issue includes a screen recording showing the exact mismatch). If a reviewer can confirm on real hardware, that would close the loop.

Reported and root-caused by @Michael-Steshenko in #5780 — credit where it's due.

[Copilot]

Copilot wasn't able to review any files in this pull request.

---

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

[Michael-Steshenko]

Thanks for the credit @stefanomainardi, I was waiting for a response from a maintainer on which of the 2 possible fixes I mention in the PR is better.
Thanks for also taking care of all other spots where this issue exists and making a clean PR.
I can validate the fix works across all the script changes when I get to my T480 tomorrow.
Would you mind adding a Co-authored-by: trailer to the commit message?

[stefanomainardi]

Done in 3d7e22f — amended with Co-authored-by: Michael Steshenko <michael.steshenko@gmail.com> (using the public email from your GitHub profile; let me know if you'd prefer the noreply alias instead and I'll swap it).

Looking forward to the T480 validation, thanks 🙌