perf: prevent cron job overlap, auto-set GOMEMLIMIT, fix tgbot userStates race

cron: SkipIfStillRunning stops a slow 5s/10s job from overlapping itself and racing the shared xrayAPI (grpc conn leak) and the StatsLastValues map (fatal concurrent map write). memlimit: auto-detect a Go soft memory limit from XUI_MEMORY_LIMIT, the cgroup limit, or system RAM (about 90 percent); opt-in pprof via XUI_PPROF. tgbot: userStates now goes through a mutex-guarded store with TTL pruning (was raced by worker-pool and delayed-delete goroutines). check_client_ip: prefilter inbounds by settings LIKE limitIp instead of loading and JSON-parsing all of them every scan. minor: prune StatsLastValues, RateLimiter.lastSent, reportedRemoteTagConflict. docker-compose: document the memory knobs.

docker-compose.yml

@@ -5,6 +5,9 @@ services:
       dockerfile: ./Dockerfile
     container_name: 3xui_app
     # hostname: yourhostname <- optional
+    # Optional hard memory cap. When set, the panel auto-derives its Go soft
+    # limit (GOMEMLIMIT, ~90%) from this so it GCs before the OOM killer fires.
+    # mem_limit: 512m
     # The bundled Fail2ban (XUI_ENABLE_FAIL2BAN below) enforces the IP limit
     # with iptables, which needs NET_ADMIN. Without these caps a ban is logged
     # and shown in fail2ban status but never actually applied. NET_RAW covers
@@ -18,6 +21,11 @@ services:
     environment:
       XRAY_VMESS_AEAD_FORCED: "false"
       XUI_ENABLE_FAIL2BAN: "true"
+      # Go memory soft limit. If neither is set, the panel auto-detects the
+      # cgroup/host limit and targets ~90%. Pin it explicitly with one of:
+      # XUI_MEMORY_LIMIT: "400"      # in MiB
+      # GOMEMLIMIT: "400MiB"         # Go syntax, takes precedence
+      # XUI_PPROF: "true"           # expose pprof on 127.0.0.1:6060 for profiling
       # XUI_INIT_WEB_BASE_PATH: "/"
       # XUI_PORT: "8080"
       # To use PostgreSQL instead of the default SQLite, run: