fix(openclaw): NEMOTRON_SYSTEM_PROMPT confidence/reasoning 移至最前

Nemo-4B 4B 參數模型輸出長度有限，confidence/reasoning 排在 schema 末尾
時常被截斷，導致 openclaw.py:1045 fallback 補 0.82 假數據。

修復：將 confidence 和 reasoning 移至 schema 最前兩個欄位，確保模型
輸出截斷時仍包含最關鍵欄位。同時明確禁止模型抄範例值。

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

apps/api/src/core/prompts.py

@@ -140,23 +140,24 @@ You are an SRE AI. Analyze the alert and respond with ONLY valid JSON.
 
 ## Required JSON Schema:
 {
+  "confidence": 0.75,
+  "reasoning": "簡短理由 (繁體中文)",
+  "primary_responsibility": "FE|BE|INFRA|DB|COLLAB",
+  "risk_level": "low|medium|critical",
   "action_title": "操作標題 (繁體中文)",
   "description": "根因分析 (繁體中文)",
   "suggested_action": "RESTART_DEPLOYMENT|DELETE_POD|SCALE_DEPLOYMENT|NO_ACTION",
   "kubectl_command": "kubectl 指令",
   "target_resource": "目標資源",
   "namespace": "K8s namespace",
-  "risk_level": "low|medium|critical",
-  "blast_radius": {"affected_pods": 1, "estimated_downtime": "~30s"},
-  "primary_responsibility": "FE|BE|INFRA|DB|COLLAB",
-  "confidence": 0.9,
-  "reasoning": "簡短理由 (繁體中文)"
+  "blast_radius": {"affected_pods": 1, "estimated_downtime": "~30s"}
 }
 
 ## Rules:
 1. Response MUST be valid JSON.
-2. Language: Traditional Chinese (Taiwan).
-3. No explanation outside JSON.
+2. confidence MUST be a real number you calculated (0.0-1.0). NOT 0.75 or 0.82.
+3. Language: Traditional Chinese (Taiwan).
+4. No explanation outside JSON.
 """
 
 PROMPT_VERSION = "7.1"