[LeonardSEO]

Experimental: not tested on a strap. This changes what the Apple 5/MG sync path trusts, so I do not expect it to be merged without a strap log confirming it.

What this PR does

The 5/MG GET_DATA_RANGE SUCCESS body carries the last-written page's clock at body offset 57 (cmdOff + 60, frame byte 70), with 0 and 0xFFFFFFFF as "no clock".

• New DataRange.pageClockNewestUnix(from:cmdOff:) (Swift + Kotlin twin) reads that field directly. Golden vectors pin a normal reply, the sentinel and a short frame on both platforms.
• Both platforms use it for 5/MG instead of the every-offset scan; WHOOP 4.0 keeps the scan.
• Apple: the 5/MG reply now passes feedsSync: true, so the newest timestamp feeds strapNewestTs and the backfill window, which Android already does today from the scan. The 4.0-only GET_CLOCK correlation (Values not computing until a week later #1598) is unaffected.

Source: the field and its confirmation come from LeonardSEO/wearable5.0 (FINDINGS §3), checked on 589 complete SUCCESS replies from 21 public capture sessions (NOOP issue attachments, Asherlc/dofek, tanarchytan/whoop-rs): the u32 at @57 is a plausible Unix time in all 589, and when the backlog is 0 it equals the trim-side clock in 286 of 286. The test vectors are wearable5.0's synthetic golden vectors, built to that layout.

Type of change

• Bug fix

How it was tested

• DataRangeTests / DataRangeScanTest: wearable5.0's golden vectors (normal reply, 0xFFFFFFFF sentinel, short frame), and pagesBehind on the same frames.
• swift test in Packages/WhoopProtocol: 817 tests pass (1 skipped, as on main).
• ./gradlew testFullDebugUnitTest: 6689 tests, all pass apart from the 2 RecoveryDriversTest half-tie cases that fail the same way on unmodified main.
• Tools/i18n_audit.py --ci, Tools/doc_comment_lint.py, docs/protocol-examples checks and Tools/parity_ledger.py --refresh-derived (derived snapshots included): pass.
• Tools unit tests, the parity acceptance suite and the R-R contract: pass apart from the 4 test_check_source_references cases (/var vs /private/var temp path on macOS) that fail the same way on unmodified main.
• Apple: the macOS app (Strand) builds with this change on my development branch; StrandTests and the iOS target were not run locally for this branch. app-build.yml covers both.

Not tested on hardware. The decode is checked against captures, not against a live 5/MG run of this build. What would confirm it: a 5/MG strap log on Apple showing Strap newest banked record: … matching the last night, and auto-continue/"Pending sync" behaving as on Android.

Checklist

• Swift package tests pass for any package I touched
• Android unit tests pass if I touched android/ (apart from the 2 RecoveryDriversTest half-tie cases that fail the same way on unmodified main)
• No new build warnings introduced (Android: none; Apple warnings not checked per branch)
• UI changes use only StrandDesign / Palette / Metrics tokens
• No hardcoded hex frame bytes; protocol facts live in the schema / decoders
• Follows the conventions in docs/CONTRIBUTING.md
• I did not commit generated output (Strand.xcodeproj/) or any secrets/keystores

Related issues

Refs #1992

🤖 Generated with Claude Code