fix(ollama): 修復容災鏈四大 bug — OFFLINE cache 放大 + SLOW 路由缺失 + recovery 命名不一致 + 告警顯示

根因：NetworkPolicy reload/CNI 瞬態抖動導致三台 Ollama 同時 OFFLINE，被 30s Redis cache 放大
  → 後續 30s 所有請求誤走 Gemini，燒 quota

B1 ollama_health_monitor: OFFLINE TTL 從 30s 縮短至 5s，儘速重試
B3 ollama_health_monitor: inference ConnectError 改判 DEGRADED（connectivity 通了不算 OFFLINE）
B5/B6 ollama_auto_recovery: _current_primary 預設改 "ollama_gcp_a"，比對改 startswith("ollama_")
SLOW 修復: failover_manager SLOW 節點視為可用（優於 Gemini quota 耗盡）
SLOW 修復: auto_recovery SLOW 也計入 recovery counter（GCP 高負載仍可切回）
告警顯示: _provider_display 加入 GCP-A/B/Local 具體伺服器識別
告警顯示: _format_automation_block 加入 Token 用量行

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

apps/api/src/services/telegram_gateway.py

@@ -218,9 +218,15 @@ class TelegramMessage:
     nemotron_latency_ms: float = 0.0  # Nemotron 呼叫延遲 (ms)
 
     def _provider_display(self) -> tuple[str, str]:
-        """Return display provider and optional model suffix."""
+        """Return display provider and optional model suffix.
+        2026-05-04 ogt: 加入具體 Ollama 伺服器顯示（GCP-A/B/Local）
+        """
         provider_names = {
             "ollama": "Ollama",
+            # 2026-05-04 ogt: ADR-110 三層容災具體伺服器識別
+            "ollama_gcp_a": "Ollama GCP-A (34.143.170.20)",
+            "ollama_gcp_b": "Ollama GCP-B (34.21.145.224)",
+            "ollama_local": "Ollama Local (111)",
             "gemini": "Gemini",
             "claude": "Claude",
             "nvidia": "Nemotron",
@@ -249,7 +255,9 @@ class TelegramMessage:
         return "safe_gate_pending"
 
     def _format_automation_block(self) -> str:
-        """Visible AI automation chain for every ACTION REQUIRED card."""
+        """Visible AI automation chain for every ACTION REQUIRED card.
+        2026-05-04 ogt: 加入 Token 用量 + 具體 Ollama 伺服器顯示
+        """
         provider_display, model_suffix = self._provider_display()
         mode = self._automation_mode()
         openclaw_state = provider_display if provider_display != "rule_fallback" else "degraded"
@@ -258,6 +266,12 @@ class TelegramMessage:
         elephant_state = "timeline_km_pending"
         flow = "webhook>investigator>router>llm/rule>safe>approval"
 
+        # 2026-05-04 ogt: Token 用量顯示（有資料才顯示）
+        token_line = ""
+        if self.ai_tokens > 0:
+            cost_str = f" / ${self.ai_cost:.4f}" if self.ai_cost > 0 else ""
+            token_line = f"├ Tokens：<code>{self.ai_tokens:,}{cost_str}</code>\n"
+
         return (
             f"🤖 <b>AI 自動化鏈路</b>\n"
             f"├ Router：<code>{html.escape(provider_display)}{model_suffix}</code>\n"
@@ -266,6 +280,7 @@ class TelegramMessage:
             f"NemoTron：<code>{html.escape(nemotron_state)}</code>\n"
             f"├ Hermes：<code>{html.escape(hermes_state)}</code> | "
             f"ElephantAlpha：<code>{html.escape(elephant_state)}</code>\n"
+            f"{token_line}"
             f"└ Flow：<code>{flow}</code>\n"
         )