Implement all vendor integrations, WAN graphs, and FortiGate health panel

- Wire up 26 vendor providers: Atlassian Statuspage API, Status.io, Instatus,
  AWS RSS feeds, Apple/Google JSON feeds, M365 Graph API, and synthetic checks
- Add 11 new providers: AWS, Cloudflare, SmartPass, School Dismissal Manager,
  SherpaDesk, Classkick, ClassDojo, Savvas, Study Island, Promethean, RAZ-Kids
- Rename Local Infrastructure → Internet (TCP check to 8.8.8.8:53)
- Add WAN throughput graph section: dual-link canvas graphs (Crown Castle +
  Comcast) polling FortiGate REST API every 30s with 30-min rolling history
- Add FortiGate health card: uptime, CPU %, memory % from FortiOS API
- Add /api/throughput and /api/fortigate-health endpoints
- Add README with setup instructions

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

README.md

@@ -0,0 +1,111 @@
+# NHSD Service Status Dashboard
+
+A web-based status feed aggregator for the North Hills School District IT department. Provides a single-pane-of-glass view of vendor service health across 26 platforms, plus live WAN throughput graphs and FortiGate health monitoring.
+
+## Features
+
+- **Vendor status cards** — real-time health for all district platforms, color-coded and sortable by severity
+- **WAN throughput graphs** — live RX/TX graphs for Crown Castle and Comcast WAN links, pulled from the FortiGate API
+- **FortiGate health panel** — hostname, firmware version, uptime, CPU, and memory utilization
+- **Auto-refresh** — vendor status refreshes every 2 minutes; throughput every 30 seconds
+- **Glanceable design** — optimized for a wall-mounted monitor; status obvious from across the room
+
+## Prerequisites
+
+- **Node.js** v18 or later
+- **Caddy** v2 — place `caddy.exe` in `bin/caddy/`
+- **NSSM** (optional, for running as Windows services) — place `nssm.exe` in `bin/nssm/`
+
+## Setup
+
+### 1. Install backend dependencies
+
+```bash
+cd backend
+npm install
+```
+
+### 2. Configure credentials
+
+Copy the template and fill in your values:
+
+```bash
+cp backend/.env.example backend/.env
+```
+
+Edit `backend/.env`:
+
+```env
+# Microsoft 365 — Azure AD app registration
+AZURE_TENANT_ID=your-tenant-id
+AZURE_CLIENT_ID=your-client-id
+AZURE_CLIENT_SECRET=your-client-secret
+
+# FortiGate — WAN throughput and system health
+FORTIGATE_HOST=10.1.20.1
+FORTIGATE_API_TOKEN=your-api-token
+FORTIGATE_WAN1_INTERFACE=port25
+FORTIGATE_WAN1_LABEL=Crown Castle
+FORTIGATE_WAN2_INTERFACE=port8
+FORTIGATE_WAN2_LABEL=Comcast
+```
+
+**Microsoft 365**: Create an app registration in Azure AD with `ServiceHealth.Read.All` permission (application, not delegated).
+
+**FortiGate**: Create a read-only API token under **System → Administrators → Create New → REST API Admin**. Only Monitor access is required — no configuration access needed.
+
+### 3. Start the backend
+
+```bash
+cd backend
+npm start
+```
+
+The backend listens on `http://localhost:3000`.
+
+### 4. Start Caddy
+
+```bash
+bin/caddy/caddy.exe run --config config/Caddyfile
+```
+
+The dashboard is then available at `https://status.nhsd.net:8443`.
+
+> **Note:** There is a separate, pre-existing Caddy instance on this machine. Always use the `--config config/Caddyfile` flag to target the dashboard Caddy specifically.
+
+## Running as Windows Services (NSSM)
+
+Use the helper scripts in `scripts/` to install the backend and Caddy as NSSM services so they start automatically with Windows.
+
+## API Endpoints
+
+| Endpoint | Description |
+|---|---|
+| `GET /api/status` | All vendor status (array) |
+| `GET /api/throughput` | WAN throughput history (array of 30-min data points) |
+| `GET /api/fortigate-health` | FortiGate hostname, version, uptime, CPU, memory |
+| `GET /api/health` | Backend liveness check |
+
+## Adding a New Vendor
+
+1. Create `backend/providers/<vendor>.js` exporting `name`, `url`, and `checkStatus()`
+2. Add it to `backend/providers/index.js`
+
+See any existing provider for the expected return shape:
+
+```js
+export const name = "Vendor Name";
+export const url  = "https://status.vendor.com/";
+
+export async function checkStatus() {
+  // ...
+  return {
+    name,
+    status,      // "operational" | "degraded" | "outage" | "unknown"
+    message,     // short human-readable description
+    lastUpdated: new Date().toISOString(),
+  };
+}
+```
+
+If a vendor check throws, the backend catches it and returns `status: "unknown"` — the dashboard degrades gracefully.