Sync workstation helpers and preserve baseline deployment

systemd/user/llama-qwen.service

@@ -1,12 +1,19 @@
 [Unit]
 Description=Local llama.cpp Qwen model server
-After=network-online.target
+After=network-online.target graphical-session.target
+StartLimitIntervalSec=300
+StartLimitBurst=3
 
 [Service]
 Type=simple
 ExecStart=%h/.local/libexec/hermes-stack-llama-run
 Restart=on-failure
-RestartSec=10
+RestartSec=30
+Nice=10
+CPUQuota=1600%
+MemoryHigh=40G
+MemoryMax=48G
+OOMPolicy=stop
 TimeoutStopSec=60
 
 [Install]