perf: optimize app registry performance with caching (calcom#22632)

* feat: optimize app registry performance with memory-cache

- Implement in-memory caching for getInstallCountPerApp with 5-minute TTL
- Add memory-cache and @types/memory-cache dependencies to packages/lib
- Remove invalidateInstallCountCache functionality as requested
- Maintain existing API compatibility for app sorting functionality
- Improve performance by avoiding expensive SQL COUNT queries on every request

Co-Authored-By: anik@cal.com <adhabal2002@gmail.com>

* update

* Update getInstallCountPerApp.ts

---------

Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com>

Author: anikdhabal

packages/lib/apps/getInstallCountPerApp.ts

@@ -1,8 +1,9 @@
+import { unstable_cache } from "next/cache";
 import { z } from "zod";
 
 import prisma from "@calcom/prisma";
 
-const getInstallCountPerApp = async () => {
+const computeInstallCountsFromDB = async (): Promise<Record<string, number>> => {
   const mostPopularApps = z.array(z.object({ appId: z.string(), installCount: z.number() })).parse(
     await prisma.$queryRaw`
     SELECT
@@ -24,4 +25,12 @@ const getInstallCountPerApp = async () => {
   }, {} as Record<string, number>);
 };
 
+const getInstallCountPerApp = async (): Promise<Record<string, number>> => {
+  return unstable_cache(async () => computeInstallCountsFromDB(), ["app-install-counts"], {
+    revalidate: 300,
+    tags: ["app-install-counts"],
+  })();
+};
+
 export default getInstallCountPerApp;
+export { computeInstallCountsFromDB };