perf(database): index WorkerDeployment on (environmentId, status, id) for the deployments list

The deployments list query filters by status and paginates by id desc, but no
index covered status, so Postgres seq-scanned the environment's deployments and
discarded ~350 non-matching rows per row returned. Add (environmentId, status,
id) so the status filter is index-satisfied and the scan is bounded to the page.

Author: ericallam

.server-changes/worker-deployment-list-status-index.md

@@ -0,0 +1,6 @@
+---
+area: webapp
+type: improvement
+---
+
+Loading the deployments list is now faster, especially when filtering by deployment status on projects with many deployments.

internal-packages/database/prisma/migrations/20260812130000_add_worker_deployment_environment_id_status_id_index/migration.sql

@@ -0,0 +1 @@
+CREATE INDEX CONCURRENTLY IF NOT EXISTS "WorkerDeployment_environmentId_status_id_idx" ON "public"."WorkerDeployment"("environmentId", "status", "id");

internal-packages/database/prisma/schema.prisma

@@ -2214,6 +2214,7 @@ model WorkerDeployment {
   @@unique([environmentId, version])
   @@index([commitSHA])
   @@index([environmentId, createdAt])
+  @@index([environmentId, status, id])
 }
 
 enum WorkerDeploymentStatus {