nutrition: pre-download HuggingFace models at build time

The deployment server couldn't fetch transformer models at runtime due to
restricted network access and permission errors writing to node_modules.
Add a prebuild script to download models during build and document
TRANSFORMERS_CACHE env var for configuring a shared writable cache path.

scripts/download-models.ts

@@ -0,0 +1,17 @@
+/**
+ * Pre-downloads HuggingFace transformer models so they're cached for runtime.
+ * Run with: pnpm exec vite-node scripts/download-models.ts
+ */
+import { pipeline } from '@huggingface/transformers';
+
+const MODELS = [
+	'Xenova/all-MiniLM-L6-v2',
+	'Xenova/multilingual-e5-small',
+];
+
+for (const name of MODELS) {
+	console.log(`Downloading ${name}...`);
+	const p = await pipeline('feature-extraction', name, { dtype: 'q8' });
+	await p.dispose();
+	console.log(`  done`);
+}