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/lib/pages.ts

@@ -0,0 +1,12 @@
+// src/lib/pages.ts
+// Sajtens riktiga sidor, härledda från src/pages/*.astro (404 och _-filer exkluderas).
+// Nya sidor hamnar automatiskt i sitemap.
+const pageFiles = import.meta.glob('../pages/*.astro');
+
+export function sitePaths(): string[] {
+  return Object.keys(pageFiles)
+    .map((file) => file.replace(/^\.\.\/pages\//, '').replace(/\.astro$/, ''))
+    .filter((name) => name !== '404' && !name.startsWith('_') && !name.startsWith('['))
+    .map((name) => (name === 'index' ? '/' : `/${name}`))
+    .sort((a, b) => (a === '/' ? -1 : b === '/' ? 1 : a.localeCompare(b)));
+}