🤖 fix: respect HTTP_PROXY env vars for network requests (#871)

Switch from undici `Agent` to `EnvHttpProxyAgent` to automatically
respect `HTTP_PROXY`, `HTTPS_PROXY`, and `NO_PROXY` environment
variables.

**Benefits:**
- Enables debugging/inspection via proxy tools like mitmproxy
- Supports corporate network environments that require proxying

The `EnvHttpProxyAgent` inherits all the same timeout configuration
(`bodyTimeout: 0`, `headersTimeout: 0`) to prevent `BodyTimeoutError` on
long-running reasoning model pauses.

_Generated with `mux`_

Author: ibetitsmike

src/node/services/aiService.ts

@@ -44,19 +44,21 @@ import type {
 } from "@/common/types/stream";
 import { applyToolPolicy, type ToolPolicy } from "@/common/utils/tools/toolPolicy";
 import { MockScenarioPlayer } from "./mock/mockScenarioPlayer";
-import { Agent } from "undici";
+import { EnvHttpProxyAgent, type Dispatcher } from "undici";
 
 // Export a standalone version of getToolsForModel for use in backend
 
 // Create undici agent with unlimited timeouts for AI streaming requests.
 // Safe because users control cancellation via AbortSignal from the UI.
-const unlimitedTimeoutAgent = new Agent({
+// Uses EnvHttpProxyAgent to automatically respect HTTP_PROXY, HTTPS_PROXY,
+// and NO_PROXY environment variables for debugging/corporate network support.
+const unlimitedTimeoutAgent = new EnvHttpProxyAgent({
   bodyTimeout: 0, // No timeout - prevents BodyTimeoutError on long reasoning pauses
   headersTimeout: 0, // No timeout for headers
 });
 
 // Extend RequestInit with undici-specific dispatcher property (Node.js only)
-type RequestInitWithDispatcher = RequestInit & { dispatcher?: InstanceType<typeof Agent> };
+type RequestInitWithDispatcher = RequestInit & { dispatcher?: Dispatcher };
 
 /**
  * Default fetch function with unlimited timeouts for AI streaming.