fix: use correct config field names with underscores

According to official JSON Schema:
- model (not defaultModel) - global default model
- small_model (not smallModel) - small model for titles
- default_agent (not defaultAgent) - default agent to use

Also added mode: primary for user-facing agents.

.kilo/kilo.jsonc

@@ -4,10 +4,14 @@
   "skills": {
     "paths": [".kilo/skills"]
   },
+  "model": "qwen/qwen3.6-plus:free",
+  "small_model": "openai/llama-3.1-8b-instant",
+  "default_agent": "orchestrator",
   "agent": {
     "orchestrator": {
       "model": "ollama-cloud/glm-5",
-      "description": "Main dispatcher. Routes tasks between agents based on Issue status."
+      "description": "Main dispatcher. Routes tasks between agents based on Issue status.",
+      "mode": "primary"
     },
     "pipeline-runner": {
       "description": "Runs agent pipeline with Gitea logging",
@@ -21,19 +25,23 @@
     },
     "code": {
       "model": "ollama-cloud/qwen3-coder:480b",
-      "description": "Primary code writer. Full tool access for development tasks."
+      "description": "Primary code writer. Full tool access for development tasks.",
+      "mode": "primary"
     },
     "ask": {
       "model": "qwen/qwen3.6-plus:free",
-      "description": "Read-only Q&A agent for codebase questions."
+      "description": "Read-only Q&A agent for codebase questions.",
+      "mode": "primary"
     },
     "plan": {
       "model": "ollama-cloud/nemotron-3-super",
-      "description": "Task planner. Creates detailed implementation plans."
+      "description": "Task planner. Creates detailed implementation plans.",
+      "mode": "primary"
     },
     "debug": {
       "model": "ollama-cloud/gpt-oss:20b",
-      "description": "Bug diagnostics and troubleshooting."
+      "description": "Bug diagnostics and troubleshooting.",
+      "mode": "primary"
     }
   }
 }