Wave-2 right-sizing: flex-auth requests 145m/288Mi -> 45m/204Mi across 9 releases (FLEX-WP-0038)

Chart default request 25m/32Mi -> 5m/20Mi (user-engine, tenant-engine,
ops-warden, ops-warden-fetch, secrets-engine); overrides keep 5m CPU with memory
by 7d p95 x1.15: application-readback 24Mi, informed-decision-lifecycle 24Mi,
informed-decision-sitting 40Mi (p95 35Mi, up from 32Mi), informed-decision-t03 16Mi.
Limits unchanged.
Evidence: resource-control 2026-10-21 node budget (3d9b1bd).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 276069@bnt-lap001
Assistant-Session: 2e192b3d-b3e8-4403-94d4-d55a2c8b476c

charts/flex-auth/values.yaml

@@ -82,8 +82,8 @@ callerAuth:
 
 resources:
   requests:
-    cpu: 25m
-    memory: 32Mi
+    cpu: 5m
+    memory: 20Mi
   limits:
     cpu: 300m
     memory: 192Mi

values/application-readback.yaml

@@ -29,7 +29,7 @@ consumer:
 resources:
   requests:
     cpu: 5m
-    memory: 32Mi
+    memory: 24Mi
   limits:
     cpu: 300m
     memory: 192Mi

values/informed-decision-lifecycle.yaml

@@ -24,7 +24,7 @@ consumer:
 resources:
   requests:
     cpu: 5m
-    memory: 32Mi
+    memory: 24Mi
   limits:
     cpu: 300m
     memory: 192Mi

values/informed-decision-sitting.yaml

@@ -22,7 +22,7 @@ consumer:
 resources:
   requests:
     cpu: 5m
-    memory: 32Mi
+    memory: 40Mi
   limits:
     cpu: 300m
     memory: 192Mi

values/informed-decision-t03.yaml

@@ -21,5 +21,5 @@ consumer:
   namespace: informed-decision
   podName: informed-decision
 resources:
-  requests: {cpu: 5m, memory: 32Mi}
+  requests: {cpu: 5m, memory: 16Mi}
   limits: {cpu: 300m, memory: 192Mi}