feat: adicionar 5 novos painéis ao dashboard (MCPs, n8n, Paperclip, IA, Operações)

Expansão do dashboard de 3 para 8 páginas com dados reais do stack:
- MCPs: monitorização de 33 MCPs no gateway com ping e estado online/offline
- n8n: 14 workflows com último run, duração e falhas 24h
- Paperclip: 16 agentes operacionais, routines e issues (PostgreSQL)
- IA/Claude: visão das 3 camadas (189 skills, 72 agents, 39 MCPs, CARL)
- Operações: tickets Desk CRM por departamento + cobertura PROCs

16 ficheiros novos (3042 linhas), 3 existentes editados.
Nova dependência: pg (PostgreSQL client para Paperclip).
Audit: 0 vulnerabilidades (npm audit fix aplicado).

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

api/services/operations.ts

@@ -0,0 +1,122 @@
+/**
+ * Operations Dashboard Service — Tickets e Cobertura de PROCs
+ * @author Descomplicar® | @link descomplicar.pt | @copyright 2026
+ */
+import db from '../db.js'
+import type { RowDataPacket } from 'mysql2'
+
+// ---------------------------------------------------------------------------
+// Tipos
+// ---------------------------------------------------------------------------
+
+export interface OperationsDashboard {
+  tickets: {
+    open: number
+    high_priority: number
+    avg_response_hours: number
+    by_department: { dept: string; count: number }[]
+  }
+  procedures: {
+    total: number
+    departments: number
+    coverage: { dept: string; procs: number; total_expected: number; pct: number }[]
+  }
+}
+
+// ---------------------------------------------------------------------------
+// Dados estáticos — cobertura de PROCs por departamento
+// ---------------------------------------------------------------------------
+
+const PROC_COVERAGE = [
+  { dept: 'D1 — Comercial',         procs: 5,  total_expected: 8  },
+  { dept: 'D2 — Suporte',           procs: 3,  total_expected: 6  },
+  { dept: 'D3 — Contabilidade',     procs: 3,  total_expected: 5  },
+  { dept: 'D4 — RH',                procs: 1,  total_expected: 4  },
+  { dept: 'D5 — Marketing',         procs: 5,  total_expected: 8  },
+  { dept: 'D6 — Gestão',            procs: 8,  total_expected: 10 },
+  { dept: 'D7 — Tecnologia',        procs: 18, total_expected: 20 },
+  { dept: 'Cross-Departamental',    procs: 5,  total_expected: 6  },
+]
+
+// ---------------------------------------------------------------------------
+// Query principal
+// ---------------------------------------------------------------------------
+
+export async function getOperationsDashboard(): Promise<OperationsDashboard> {
+  // Queries paralelas para melhor performance
+  const [
+    ticketsAbertosResult,
+    ticketsAltaPrioridadeResult,
+    ticketsPorDepartamentoResult,
+    tempoMedioRespostaResult,
+  ] = await Promise.all([
+    // a) Total de tickets abertos (status 1=Open, 2=In Progress, 3=Answered)
+    db.query<RowDataPacket[]>(
+      `SELECT COUNT(*) as count
+       FROM tbltickets
+       WHERE status IN ('1','2','3')`
+    ),
+
+    // b) Tickets de alta prioridade ainda abertos (priority 2=High, 3=Urgent)
+    db.query<RowDataPacket[]>(
+      `SELECT COUNT(*) as count
+       FROM tbltickets
+       WHERE priority IN (2,3)
+         AND status IN ('1','2','3')`
+    ),
+
+    // c) Tickets abertos agrupados por departamento
+    db.query<RowDataPacket[]>(
+      `SELECT d.name as dept, COUNT(t.ticketid) as count
+       FROM tbltickets t
+       LEFT JOIN tbldepartments d ON t.department = d.departmentid
+       WHERE t.status IN ('1','2','3')
+       GROUP BY d.departmentid, d.name
+       ORDER BY count DESC`
+    ),
+
+    // d) Tempo médio de resposta em horas (últimos 90 dias)
+    db.query<RowDataPacket[]>(
+      `SELECT ROUND(AVG(TIMESTAMPDIFF(HOUR, t.date, t.lastreply)), 1) as avg_hours
+       FROM tbltickets t
+       WHERE t.lastreply IS NOT NULL
+         AND t.lastreply != '0000-00-00 00:00:00'
+         AND t.date > DATE_SUB(CURDATE(), INTERVAL 90 DAY)`
+    ),
+  ])
+
+  // Extrair valores das queries
+  const ticketsAbertos       = (ticketsAbertosResult[0][0] as RowDataPacket)?.count ?? 0
+  const ticketsAltaPrioridade = (ticketsAltaPrioridadeResult[0][0] as RowDataPacket)?.count ?? 0
+  const avgResponseHours     = (tempoMedioRespostaResult[0][0] as RowDataPacket)?.avg_hours ?? 0
+
+  const byDepartment = (ticketsPorDepartamentoResult[0] as RowDataPacket[]).map(row => ({
+    dept:  (row.dept as string) || 'Sem departamento',
+    count: Number(row.count) || 0,
+  }))
+
+  // Calcular cobertura em percentagem
+  const coverage = PROC_COVERAGE.map(item => ({
+    dept:           item.dept,
+    procs:          item.procs,
+    total_expected: item.total_expected,
+    pct:            Math.round((item.procs / item.total_expected) * 100),
+  }))
+
+  const totalProcs       = PROC_COVERAGE.reduce((sum, d) => sum + d.procs, 0)
+  const totalDepartments = PROC_COVERAGE.filter(d => d.dept.startsWith('D')).length
+
+  return {
+    tickets: {
+      open:              Number(ticketsAbertos),
+      high_priority:     Number(ticketsAltaPrioridade),
+      avg_response_hours: Number(avgResponseHours),
+      by_department:     byDepartment,
+    },
+    procedures: {
+      total:       totalProcs,
+      departments: totalDepartments,
+      coverage,
+    },
+  }
+}