Add the A4 report designer and fix band layout

Brings in the A4 (millimetre) report designer alongside the existing
matrix editor, migrates the frontend from react-scripts to Vite/Vitest,
and fixes four defects found while using the A4 band tools.

Layout — layout.js now implements the standard banded-report model:
everything prints where it was drawn, and a band growing per data row is
the only thing that moves anything. Previously `paginate` re-flowed every
band from the top margin, so a band placed halfway down the page
previewed at the top, and `header` was treated as a page header and
hoisted above the statics drawn before it. A header is a COLUMN header:
it prints in place and repeats only on a continuation page. Statics now
take part too — one drawn below a band keeps its drawn gap as that band
grows, and follows onto the next page rather than vanishing when pushed
past the bottom margin. The displacement is keyed on the band's design
bottom, not on an instance's laid-out y: every row of a band shares one
design position, so keying on the instance measured the gap from row 1
and let the static overlap rows 2..n.

Band children — a child's x is relative to its band, as its y already
was. layout() emitted it raw, so a field printed measured from the paper
edge instead of the band's left margin. Instances now carry bandX.

Band placement — a band's x comes from the page margins, so creation
takes only the click's Y (the factory's x was being overwritten by the
spread order) and dragging is locked to the vertical axis. Bands render
as a strip, with top/bottom rules only, matching the matrix editor.
fromApiReport reflows on load so reports saved with a stray x straighten
themselves. Matrix bands are constrained to vertical too.

Drag lifecycle — useElementDrag re-subscribed its document listeners on
every render during a drag, because the effect depended on an unmemoized
onDrag. Each re-subscribe removed the mouseup listener before adding the
next, measured as add/remove/add/remove/add across two mousemoves, and a
release landing in that gap was lost: the element kept following the
cursor until the next click. Snapping made it routine, since a held
guide stops x/y changing and leaves the guide overlay's render sitting
between the move and the release. onDrag and dragData now go through
refs so the effect depends on [isDragging] alone.

Note: the matrix editor has no test coverage, so its band constraint is
verified by reading only. The A4 suite is 201 passing tests.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

.claude/agent-memory/react-expert-dev/a4-entry-points.md

@@ -0,0 +1,46 @@
+---
+name: a4-entry-points
+description: How the two report editors are reached and dispatched between — kind decides the route, and why the A4 editor now needs Router + NomenclatureData context in tests
+metadata:
+  type: project
+---
+
+Phase 7 wired the A4 designer to real operator entry points. Three things here
+are easy to get wrong later.
+
+**Report `kind` decides the route, never a default.** `Main.jsx` fetches
+`/api/reports/?name=notice_report` and branches: `kind === 'a4'` → dynamic
+`import()` of `a4/print/printReport` and `printA4Report(report, data)`;
+otherwise the existing `ReportPreviewModal`. "Design Notice" routes to
+`/report-editor-a4` or `/report-editor` by the same field.
+
+**Why it matters:** a matrix report opened in the mm editor reads its character
+x/y as millimetres and looks like a corrupt layout rather than a routing bug —
+a slow, confusing failure. The A4 editor's auto-load therefore filters on
+`?kind=a4&name=...`, not name alone; verified against the real DB that
+`?kind=a4&name=<a matrix report>` returns 0 rows.
+
+**How to apply:** any new entry point that opens a saved report must resolve
+`kind` before choosing an editor or a render path. Never assume a name implies
+a kind.
+
+**`A4ReportEditor` now requires two contexts.** It calls `useLocation`/
+`useNavigate` (router-state auto-load) and `useNomenclatureData` (sample-data
+fallback). Both throw outside their providers, so any jsdom test mounting the
+whole editor needs `<MemoryRouter>` plus
+`vi.mock('../../../../contexts/NomenclatureDataContext', ...)`. Mocking the
+nomenclature hook rather than standing up the provider keeps the SSE client and
+auth context out of tests that are really about drag geometry.
+
+**Why:** adding these deps silently broke four existing `a4Guides` tests with
+"useLocation() may be used only in the context of a <Router>" — the component
+was fine, the harness was stale.
+
+**`window.print()` is not gated by transient user activation** the way
+`window.open` is, so an `await` before it is safe. The A4 print chunk is still
+prefetched on hover/focus of the Print Notice button, purely to shorten the gap
+on a slow station. This was reasoned, not verified in a real browser — if Print
+Notice is ever reported as doing nothing on click, re-test that assumption
+first.
+
+Related: [[a4-print-path]], [[a4-designer-conventions]], [[testing-react]]