feat(seo): site config, sitemap, robots.txt, llms.txt and a real 404 page

- astro.config: site, trailingSlash "never", build.format "file".
- /sitemap-index.xml + /sitemap-0.xml as prerendered endpoints (no new
  dependency); pages are derived from src/pages/*.astro, 404 excluded.
- robots.txt: all bots allowed, AI crawlers/training bots explicitly
  allowed (Linus's decision), /download/ disallowed, sitemap line.
- /llms.txt generated from business.ts, incl. usage-with-attribution note.
- 404.astro in Swedish with links back, noindex.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WvE3C43Uwtv2vtnoFBLDw7

src/pages/sitemap-0.xml.ts

@@ -0,0 +1,23 @@
+// src/pages/sitemap-0.xml.ts -> /sitemap-0.xml
+// Listar sajtens riktiga sidor (inte 404). Ingen <lastmod>: vi har inget
+// pålitligt ändringsdatum per sida, och ett påhittat datum gör mer skada än nytta.
+import type { APIRoute } from 'astro';
+import { sitePaths } from '../lib/pages';
+
+export const prerender = true;
+
+const escapeXml = (s: string) =>
+  s.replace(/&/g, '&amp;').replace(/</g, '&lt;').replace(/>/g, '&gt;').replace(/"/g, '&quot;');
+
+export const GET: APIRoute = ({ site }) => {
+  const base = site ?? new URL('https://imagesbyolofsson.se');
+  const urls = sitePaths()
+    .map((p) => `  <url><loc>${escapeXml(new URL(p, base).toString())}</loc></url>`)
+    .join('\n');
+  const body = `<?xml version="1.0" encoding="UTF-8"?>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+${urls}
+</urlset>
+`;
+  return new Response(body, { headers: { 'Content-Type': 'application/xml; charset=utf-8' } });
+};