docs: catch CLAUDE.md and the UI review up with the shared-expenses rebuild

The redesign is live (2026-07-28); only the loan model remains a proposal.
The UI review's 'settled is dead data' guidance described the pre-rebuild
state and is superseded — settled now gates ACTIVE_OBLIGATION.

CLAUDE.md

@@ -506,11 +506,12 @@ See `README.md` → **Known Gaps / TODOs** for full details.
 
 **Payment provider tracking**: `merchant_normalized` currently conflates payment provider (PayPal, Afterpay, Zip) with the actual merchant. Plan: add `payment_provider` column, update Gemini prompt to extract it separately, backfill from `merchant_name` patterns, surface in UI filters.
 
-### Open as of 2026-07-26
+### Open as of 2026-07-29
 
-- **Shared expenses redesign** — `docs/shared-expenses-design.md`. Phase 0 done;
-  Phases 1–4 unbuilt. Deliberately paused to live with the current behaviour
-  before committing to a model designed in one session.
+- **Shared expenses: loan section only** — the redesign in
+  `docs/shared-expenses-design.md` is built and live as of 2026-07-28; the
+  loan model at the end of that doc remains a proposal (Sonu's `…emi`
+  contributions still read as ordinary transfers).
 - **Expense baseline / emergency reserve** — `docs/expense-baseline.md`. One-off
   analysis, nothing built. Records four data corrections the raw numbers need
   (misfiled Raiz/super/brokerage debits, `other` credits read as negative spend,

docs/ui-information-architecture-review.md

@@ -530,9 +530,10 @@ The unfiltered balance cards correctly implement a running ledger:
 
 > splits minus payments
 
-This is coherent and should not be changed to exclude splits marked `settled`.
-The `transaction_splits.settled` field is dead data and must not be used for new
-UI claims.
+*(Superseded 2026-07-28: the shared-expenses rebuild made `settled`
+load-bearing — it gates `ACTIVE_OBLIGATION` and survives split rewrites. See
+`docs/shared-expenses-design.md` for the live model; the paragraph above
+described the pre-rebuild state.)*
 
 ### What the current model cannot answer