feat: split photo-import into vision recognition + text generation

The photo-import flow used one vision-capable model to both read the
photo and structure the full recipe (quantities, steps, timing) in a
single call. Split it into two: a vision model recognizes what's in
the picture, then a text model reconstructs the recipe from that
description — same pattern the pantry photo-scan already uses for
recognition, and lets structuring use whichever model is actually
configured for text generation. Still counts as one AI-quota unit.

v0.37.0

apps/web/lib/changelog.ts

@@ -1,5 +1,5 @@
 // Mirrors CHANGELOG.md at the repo root — update both together.
-export const APP_VERSION = "0.36.0";
+export const APP_VERSION = "0.37.0";
 
 export type ChangelogEntry = {
   version: string;
@@ -11,6 +11,13 @@ export type ChangelogEntry = {
 };
 
 export const CHANGELOG: ChangelogEntry[] = [
+  {
+    version: "0.37.0",
+    date: "2026-07-17 12:30",
+    fixed: [
+      "Generating a recipe from a photo now uses two separate AI steps — a vision model recognizes what's in the picture, then a text model writes the recipe from that — instead of one model doing both, so each step can use the model actually suited to it.",
+    ],
+  },
   {
     version: "0.36.0",
     date: "2026-07-17 12:00",