docs: fix config examples to match config.go structs

- providers.md: agents.defaults.model → model_name (4 occurrences)
- providers.md: remove non-existent session.backlog_limit field
- credential_encryption.md: add required model field to model_list example
- ANTIGRAVITY_AUTH.md: agents.defaults.model → model_name

docs/ANTIGRAVITY_AUTH.md

@@ -584,7 +584,7 @@ Each SSE message (`data: {...}`) is wrapped in a `response` field:
   ],
   "agents": {
     "defaults": {
-      "model": "gemini-flash"
+      "model_name": "gemini-flash"
     }
   }
 }

docs/credential_encryption.md

@@ -30,6 +30,7 @@ enc://AAAA...base64...
   "model_list": [
     {
       "model_name": "gpt-4o",
+      "model": "openai/gpt-4o",
       "api_key": "enc://AAAA...base64...",
       "api_base": "https://api.openai.com/v1"
     }

docs/providers.md

@@ -95,7 +95,7 @@ This design also enables **multi-agent support** with flexible provider selectio
   ],
   "agents": {
     "defaults": {
-      "model": "gpt-5.4"
+      "model_name": "gpt-5.4"
     }
   }
 }
@@ -268,7 +268,7 @@ The old `providers` configuration is **deprecated** but still supported for back
   ],
   "agents": {
     "defaults": {
-      "model": "glm-4.7"
+      "model_name": "glm-4.7"
     }
   }
 }
@@ -300,7 +300,7 @@ This keeps the runtime lightweight while making new OpenAI-compatible backends m
   "agents": {
     "defaults": {
       "workspace": "~/.picoclaw/workspace",
-      "model": "glm-4.7",
+      "model_name": "glm-4.7",
       "max_tokens": 8192,
       "temperature": 0.7,
       "max_tool_iterations": 20
@@ -330,12 +330,11 @@ picoclaw agent -m "Hello"
 {
   "agents": {
     "defaults": {
-      "model": "anthropic/claude-opus-4-5"
+      "model_name": "anthropic/claude-opus-4-5"
     }
   },
   "session": {
-    "dm_scope": "per-channel-peer",
-    "backlog_limit": 20
+    "dm_scope": "per-channel-peer"
   },
   "providers": {
     "openrouter": {