fix: Use server-side timestamps for agent polling metrics

Replaces agent-supplied timestamps with server-side DateTime.utc_now()
to prevent clock drift issues between agents from causing weird graphs.

All metrics now use consistent server time regardless of polling method:
- Agent polling (via WebSocket): server time on receipt
- Phoenix polling (direct SNMP): server time on poll (already correct)

This ensures time-series data aligns correctly when graphing metrics
from devices polled by different agents with potentially different
system clocks.

lib/towerops_web/channels/agent_channel.ex

@@ -914,7 +914,8 @@ defmodule ToweropsWeb.AgentChannel do
   defp process_polling_result(device, result, _socket) do
     snmp_device = device.snmp_device
     oid_values = Map.new(result.oid_values)
-    timestamp = DateTime.from_unix!(result.timestamp, :second)
+    # Use server time to avoid clock drift issues between agents
+    timestamp = DateTime.utc_now()
 
     Logger.info(
       "Processing polling result for #{device.name}: #{map_size(oid_values)} OIDs, #{length(snmp_device.sensors)} sensors, #{length(snmp_device.interfaces)} interfaces"