Loadpoint: correct phase sync when measured phases fall below commanded

[andig]

fixes #31491

The measured-phase fallback in syncCharger only corrected the mismatch in one direction: when the charger reports more active phases than expected. It never handled the opposite case, so when a 1p→3p switch is commanded but never physically takes effect (relay/vehicle doesn't actually engage the third phase), measuredPhases sticks at the lower value. Since maxActivePhases() is capped by measuredPhases, this permanently blocks any further scale-up attempt for the rest of the session, matching the reported symptom of 1p/3p switching "getting stuck" until a restart.

• syncCharger now corrects lp.phases whenever the measured phase count differs from the commanded one (not just when it's higher), restoring the self-healing loop that lets pvScalePhases retry 3p later
• Updated TestSyncChargerPhasesByMeasurement to assert the corrected behavior

🤖 Generated with Claude Code

[sourcery-ai]

Hey - I've found 1 issue, and left some high level feedback:

• The warning message "charger logic error: phases mismatch" is now triggered for both over- and under-reported phases; consider rewording it or adding more context so it accurately reflects both directions of mismatch.
• The new condition chargerPhases > 0 && chargerPhases != phases intentionally ignores a zero-measured-phases case; if a 3p→0p failure is possible in practice, it might be worth clarifying or revisiting why 0 is excluded here.

Prompt for AI Agents

Please address the comments from this code review:

## Overall Comments
- The warning message "charger logic error: phases mismatch" is now triggered for both over- and under-reported phases; consider rewording it or adding more context so it accurately reflects both directions of mismatch.
- The new condition `chargerPhases > 0 && chargerPhases != phases` intentionally ignores a zero-measured-phases case; if a 3p→0p failure is possible in practice, it might be worth clarifying or revisiting why 0 is excluded here.

## Individual Comments

### Comment 1
<location path="core/loadpoint_sync_test.go" line_range="188-189" />
<code_context>
 		{1, 1, 1},
 		{1, 3, 3},
 		{3, 0, 3},
-		{3, 1, 3}, // ignore
+		{3, 1, 1}, // force
 		{3, 3, 3},
 	}
</code_context>
<issue_to_address>
**suggestion (testing):** Clarify and extend tests around zero measured phases vs non-zero mismatches

With `chargerPhases > 0 && chargerPhases != phases`, `{3, 0, 3}` now shows that a 0 measurement is ignored, while `{3, 1, 1}` covers non-zero mismatches that trigger correction. To make this clearer and ensure both directions are tested, consider (a) updating the `{3, 0, 3}` comment to `// ignore 0 measurement`, and (b) adding `{1, 3, 3}` with `// force up`, or a dedicated test, to document the symmetric low→high behavior if it’s handled by the same logic.
</issue_to_address>

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}