diag: show repeat counts in the mDNS query log; bump prop submodule

[jonasbark]

Why

Tester feedback on 6.3.0+22 (the build with the QU fix from #372): still failing, tnc still required. Their diagnostics changed the diagnosis.

The mDNS exchange is working end-to-end

20:32:38 ... QU  PTR _openbikecontrol._tcp.local              → unicast+multicast
20:32:38 ... QM  SRV BikeControl._openbikecontrol._tcp.local  → multicast
20:32:38 ... QM  TXT BikeControl._openbikecontrol._tcp.local  → multicast

MyWhoosh could only ask for SRV BikeControl._openbikecontrol._tcp.local if it had already received and parsed our PTR answer — that instance name came from us. The browse succeeded. It then got SRV and TXT, and our SRV response carries the A record as an additional, so it had PTR + SRV + TXT + A and still showed no dialog until tnc.

So the shared-port/QU fix is not what blocks this tester. It may be why the browse worked here at all — before it, all three interface copies would have been unicast-only — but that counterfactual isn't provable from this capture. The failure is downstream of discovery, consistent with MyWhoosh resolving the SRV target through the Windows resolver (GetAddrInfo) rather than using the A record we hand it. Tailscale is present on this machine (100.111.140.105), which is the known way .local resolution breaks.

The log couldn't answer the remaining question

No A-record query appears anywhere in the capture — the single most informative fact — and I can't tell whether none was sent or the window was simply too short. It covered eight seconds; the tnc at 20:33:30 and the connect at 20:33:35 had aged out.

Changes

Renders ×N on folded entries so a continuous poller reads as one line that fired N times. Single occurrences stay unmarked. Picks up the prop-side fold + ceiling raise (OpenBikeControl/prop#2).

Also visible in the capture — not a code change

MyWhoosh browses for both services and we answer only one:

20:32:39 ... QM PTR _openbikecontrol._tcp.local    → multicast
20:32:39 ... QM PTR _wahoo-fitness-tnp._tcp.local  → no answer

This tester is on the OpenBikeControl method, so we advertise _openbikecontrol._tcp only. Worth having them retry with the Wahoo/Network connection method — MyWhoosh's native trainer path, which may not depend on the same resolve step.

Tests

prop 487/487, app services 64/64, flutter analyze clean on both changed files.

🤖 Generated with Claude Code

https://claude.ai/code/session_01PanTG9kRBi4QTQ2psvpTxP

[Copilot]

Pull request overview

Improves the diagnostics output for the “mDNS queries received” section by making folded/repeated mDNS queries visible via an inline repeat-count marker, and adds tests to lock in the new formatting behavior.

Changes:

• Append a ×N repeat-count marker to folded mDNS query log entries when count > 1.
• Add unit tests verifying repeat-count rendering and ensuring single occurrences remain unmarked.

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated no comments.

File	Description
lib/services/debug_diagnostics.dart	Appends ×N to repeated (folded) mDNS query log lines when count > 1.
test/services/debug_diagnostics_test.dart	Adds coverage for repeat-count marker presence/absence in the diagnostics text output.

---

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