fix: cap Prisma pool during next build to stop deploy SSG timeouts

Next build workers were each opening up to DATABASE_POOL_SIZE connections and exhausting MySQL (pool active=0), hanging sitemap generation. Cap build pool to 5, fail connect faster, limit SSG concurrency, and make sitemap dynamic.

Co-authored-by: Cursor <[email protected]>

src/app/sitemap.ts

@@ -1,6 +1,10 @@
 import type { MetadataRoute } from "next";
 import { prisma } from "@/lib/prisma";
 
+// Built at request time — avoids competing with SSG workers for scarce DB
+// connections during `next build` (pool timeouts killed deploy on sitemap).
+export const dynamic = "force-dynamic";
+
 export default async function sitemap(): Promise<MetadataRoute.Sitemap> {
 	const appUrl = process.env.APP_URL ?? "http://localhost:3000";