feat(mcp): payroll e2e parity: staged salary-run booking + absence deletion (#1075)

* feat(mcp): payroll e2e parity: staged salary-run booking + absence deletion

Close the last MCP-surface gaps for running payroll end-to-end via the
connector (the v1 REST API already had the full chain):

- gnubok_book_salary_run: stages a high-risk book operation; on approval
  the executor walks review -> approved -> paid -> booked via the new
  lib/salary/book-run.ts (extracted from the dashboard book route, which
  now calls the same core) and posts the immutable salary vouchers.
- gnubok_delete_absence: staged inverse of gnubok_register_absence,
  reusing deleteAbsenceRange with a dry-run day-count preview.
- Wire the missing payroll operation types into the Granskning label map
  (register_absence, update_payslip_line, employee ops, vacation_year_close
  had translations but fell back to humanized snake_case).
- Update stale 'booking happens in the web UI' prose in tool descriptions,
  the payroll-monthly skill, and the workflow hint; payload-size ceiling
  56K -> 57K per the documented bump protocol.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

* fix(mcp): widen pending_operations op-type CHECK + roster typing for book_salary_run

The op-type audit (pg-real) caught the exact bug class it exists for:
book_salary_run and delete_absence were staged in code without the
constraint-expansion migration, so every real staging INSERT would have
failed with check_violation while dry_run previewed clean. Ships the
documented widen (NOT VALID) + validate migration pair. Also fixes the
strict-mode cast in book-run.ts that failed the production typecheck.

Verified locally against supabase/postgres 15.8.1.060 with all migrations
applied: op-type audit green, pg-real 692/693 (the one failure is the
pre-existing TZ-sensitive get_unlinked_1930_lines assertion, green under
TZ=UTC as in CI).

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

---------

Co-authored-by: Claude Fable 5 <noreply@anthropic.com>

DECISIONS.md

@@ -221,3 +221,6 @@ One line per decision: `[YYYY-MM-DD] <decision>: <why>`. Appended by agents and
 [2026-07-17] EU-trade/PS settings stay opt-in flags; a ledger-derived signal (postings on 3108/3308/3107, 15 months) only renders a suggestion callout in tax settings. Auto-flipping registration flags from ledger data would assert a Skatteverket registration we cannot know.
 [2026-07-19] Voucher-sequence resync run on prod via execute_sql BEFORE the migration merges: data-only idempotent DML (no schema_migrations orphan risk) and a user was hard-blocked on year-end; migration file 20260719100000 ships the same SQL so every environment replays it as a no-op.
 [2026-07-20] Menu rename Fakturor -> Kundfakturor scoped to nav label + /invoices page title + command palette only (support tip 2026-07-19): API-key scope labels, AR-ledger xlsx sheet name, and customer/supplier detail section headings keep "Fakturor" since their surrounding context already disambiguates, and renaming API scope labels would churn a stable admin surface.
+[2026-07-20] MCP gnubok_book_salary_run walks the whole review->approved->paid->booked chain in ONE staged op instead of mirroring the dashboard's four separate clicks: the human approval of the pending operation (high-risk, confirmed=true) IS the authorization act, and a four-op chain over MCP would just be approval theater. Missing bank details downgrade from overridable block to warnings (dashboard force-approve semantics): the payment-file generators hard-block on them where it matters.
+[2026-07-20] No gnubok_archive_employee tool: gnubok_update_employee already takes is_active=false (soft-archive, BFL retention) and the v1 REST surface has the DELETE verb; a dedicated tool would only bloat the tools/list budget.
+[2026-07-20] Booking core extracted to lib/salary/book-run.ts and shared by the dashboard route + book_salary_run executor; the v1 book route intentionally keeps its own strict-mode mirror (optimistic locking, period pre-check, its own envelope) rather than being folded in.