Fix GeoPos DMS edge carry rollback at clamped degree limits

[Copilot]

Brief description of what you did

This addresses the review comment on ui_geomap.cpp where DMS wrap-carry could cause discontinuities at clamped latitude/longitude bounds (e.g., 90°59'59" incrementing to 90°00'00").
The change keeps normal digit carry behavior, while rolling back wrapped minutes/seconds when the carry target is clamped and unchanged, so users can still fully increment/decrement digits across limits and around zero transitions.

• Issue scope

  • Only the referenced wrap callbacks in GeoPos were updated (lat/lon seconds→minutes and minutes→degrees).

• Behavior change

  • On wrap, capture pre-carry value.
  • Attempt carry into next field.
  • If carry target did not change (clamped), undo the wrap for the current field (59 on increment, 0 on decrement).

• Result

  • Eliminates edge discontinuities at max/min degree bounds.
  • Preserves expected carry/borrow progression in non-clamped paths.

const auto old_minutes = field_lat_minutes.value();
field_lat_minutes.on_encoder(v);
if (field_lat_minutes.value() == old_minutes) {
    field_lat_seconds.set_value((v > 0) ? 59 : 0);
}

---

Proof that your changes work

N/A in this write-up.

🖥️ Proof it compiles

N/A in this write-up.

📱 Proof of testing on a real device

N/A in this write-up.

📡 Proof against a real emitter/receiver (if applicable)

N/A — UI numeric wrap/carry logic change; no RF path modified.

---

📚 Wiki documentation commitment

• I will (or already) create(d) wiki document for my newly added feature

---

Checklist

• Kept changes minimal and limited to necessary files
• Verified functionality remains intact and code compiles
• Attached proof that the code compiles successfully (or marked N/A as a trusted contributor)
• Attached proof of testing on real PortaPack hardware (or marked N/A as a trusted contributor)
• Attached proof of testing against a real emitter/receiver (if RF-related), or marked N/A with justification
• I understand that by getting this PR merged, I am implicitly agreeing to create or update the corresponding wiki page (including a main-screen screenshot, description, controls, and limitations)
• I own all rights to this code (i.e., all code contained in this PR), including compliant usage rights for third-party libraries, and I agree that this code is licensed under the license of this project (GPL-3.0).
• If any third-party libraries are used, I confirm that their licenses comply with the requirements for contributing to this repository.
• Reviewed the Contributing Guidelines

[gullradriel]

@copilot confirm if the conditions are there that going up/down seconds can wrap/increment/decrement minutes and hours too

[Copilot]

@copilot confirm if the conditions are there that going up/down seconds can wrap/increment/decrement minutes and hours too

Yes — the current logic keeps normal carry/borrow from seconds → minutes → degrees in both directions, and only blocks the rollover when the degree limit is already clamped.

Confirmed cases:

• 12°59'59" +1s → 13°00'00"
• 13°00'00" -1s → 12°59'59"
• 0°00'00" -1s → -1°59'59"
• 89°59'59" +1s → 90°00'00"
• At the max edge, 90°59'59" +1s stays 90°59'59" (no discontinuity).

[gullradriel]

good to go