[LeonardSEO]

What this PR does

Adds a Sleep planner card to the Sleep tab on Android and macOS/iOS: tonight's sleep target at three goals (70 / 85 / 100 % of tonight's need), the time in bed each one takes at the user's typical efficiency, and the bedtime that implies for the next wake.

• Tonight's need = the base need and debt the existing Hours-vs-Needed figures use, minus today's naps. Naps only repay debt, so they never push tonight below the base need.
• Wake time = the armed smart alarm's next fire (through the same resolver the strap alarm is armed from), else last night's wake, else 07:00. The card names which one it used.
• Time in bed = target / median efficiency of the last 14 nights (clamped to 50-100 %). With no efficiency history it equals the target, so no awake allowance is invented.
• No plan until there are three consecutive nights with sleep.
• Pure math in SleepPlanner (StrandAnalytics) with a Kotlin twin and mirrored fixtures. Display-only: nothing feeds a score, a baseline or a stored value.
• A new reorderable Sleep section with the raw key sleepPlanner on both platforms (not in the .noopbak whitelist, like the other section keys).

The 70/85/100 ladder and the three-night gate follow the shape of the official app's sleep planner (seen in its shipped resources; facts only, no code or strings). The goal names are NOOP's own: Minimum / Solid / Full.

Type of change

• New feature

How it was tested

• SleepPlannerTests / SleepPlannerTest (same fixtures on both platforms: three-night gate, nap credit only repaying debt, efficiency median and clamp, bedtime wrap past midnight); SleepLayoutPrefsTests / SleepLayoutPrefsTest and SleepSectionReorderTest updated for the new section.
• swift test in Packages/StrandAnalytics: 2149 tests pass.
• ./gradlew testFullDebugUnitTest: 6697 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.

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

None.

🤖 Generated with Claude Code