feat(headroom): make the compression request timeout configurable

The 3000 ms timeout on /v1/compress was fixed, so busy or slow machines
timed out often and sent the LLM an inconsistently compressed body,
hurting prompt caching. Add a headroomTimeoutMs setting, thread it from
the chat handler down to compressWithHeadroom, expose it in the Token
Saver dashboard, and normalize invalid values back to the 3000 ms default.

src/sse/handlers/chat.js

@@ -274,6 +274,7 @@ async function handleSingleModelChat(body, modelStr, clientRawRequest = null, re
       headroomEnabled: !!chatSettings.headroomEnabled,
       headroomUrl: chatSettings.headroomUrl || DEFAULT_HEADROOM_URL,
       headroomCompressUserMessages: !!chatSettings.headroomCompressUserMessages,
+      headroomTimeoutMs: chatSettings.headroomTimeoutMs,
       cavemanEnabled: !!chatSettings.cavemanEnabled,
       cavemanLevel: chatSettings.cavemanLevel || "full",
       ponytailEnabled: !!chatSettings.ponytailEnabled,