feat: skip tracing for health and metrics endpoints

Health check and metrics endpoints are frequently polled and don't
provide meaningful trace data. This change skips OpenTelemetry span
creation for:
- /health/* (liveness, readiness checks)
- /metrics (Prometheus metrics)

These endpoints still record Prometheus metrics (request count, latency,
in-flight requests) but no longer create trace spans, reducing tracing
noise and storage costs.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

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

nextcloud_mcp_server/observability/middleware.py

@@ -66,7 +66,11 @@ class ObservabilityMiddleware(BaseHTTPMiddleware):
         # Record start time
         start_time = time.time()
 
+        # Skip tracing for health/metrics endpoints to reduce noise
+        should_trace = not (path.startswith("/health/") or path == "/metrics")
+
         try:
+            if should_trace:
                 # Create span for request (OpenTelemetry auto-instrumentation will create parent span)
                 with trace_operation(
                     f"HTTP {method} {endpoint}",
@@ -92,6 +96,20 @@ class ObservabilityMiddleware(BaseHTTPMiddleware):
                         duration=duration,
                     )
 
+                    return response
+            else:
+                # No tracing for health/metrics endpoints, but still record metrics
+                response = await call_next(request)
+
+                # Record metrics
+                duration = time.time() - start_time
+                self._record_request_metrics(
+                    method=method,
+                    endpoint=endpoint,
+                    status_code=response.status_code,
+                    duration=duration,
+                )
+
                 return response
 
         except Exception: