remove: HF token settings section and JS fallback

ZeroGPU quota attribution works automatically via the HF JWT session
(confirmed in HAR — x-zerogpu-token/uuid headers present on all regen
calls). Manual token input is redundant.

app.py

@@ -2272,16 +2272,7 @@ _GLOBAL_JS = """
         for (var k in zerogpuHeaders) { regenHeaders[k] = zerogpuHeaders[k]; }
         console.log('[fireRegen] using zerogpu headers from parent frame');
       } else {
-        // Fallback: try user-supplied HF token from Settings
-        var userHfToken = '';
-        var hfTokenEl = document.getElementById('hf_token_input');
-        if (hfTokenEl) { var inp = hfTokenEl.querySelector('input,textarea'); if (inp) userHfToken = (inp.value || '').trim(); }
-        if (userHfToken) {
-          regenHeaders['Authorization'] = 'Bearer ' + userHfToken;
-          console.log('[fireRegen] using HF token fallback for quota attribution');
-        } else {
-          console.warn('[fireRegen] no zerogpu headers or HF token — may use anonymous quota');
-        }
+        console.warn('[fireRegen] no zerogpu headers available — may use anonymous quota');
       }
       fetch('/gradio_api/queue/join', {
         method: 'POST',
@@ -2484,19 +2475,6 @@ _GLOBAL_JS = """
 """
 
 with gr.Blocks(title="Generate Audio for Video", css=_SLOT_CSS, js=_GLOBAL_JS) as demo:
-    with gr.Accordion("⚙️ Settings (optional)", open=False):
-        gr.Markdown(
-            "**HF Token** — Paste your [Hugging Face token](https://huggingface.co/settings/tokens) here so that "
-            "segment regeneration uses your own ZeroGPU quota (PRO = 25 min/day) instead of the shared free pool. "
-            "The token is stored only in your browser and never sent anywhere except the Space's own API."
-        )
-        hf_token_input = gr.Textbox(
-            label="Hugging Face Token (optional)",
-            placeholder="hf_...",
-            type="password",
-            elem_id="hf_token_input",
-            max_lines=1,
-        )
     gr.Markdown(
         "# Generate Audio for Video\n"
         "Choose a model and upload a video to generate synchronized audio.\n\n"
@@ -2757,10 +2735,9 @@ with gr.Blocks(title="Generate Audio for Video", css=_SLOT_CSS, js=_GLOBAL_JS) a
         api_name="xregen_hunyuan",
     )
 
-    # NOTE: ZeroGPU quota attribution is now handled via postMessage("zerogpu-headers")
+    # NOTE: ZeroGPU quota attribution is handled via postMessage("zerogpu-headers")
     # to the HF parent frame — the same mechanism Gradio's own JS client uses.
     # This replaced the old x-ip-token relay approach which was unreliable.
-    # The HF token input in Settings is kept as a fallback.
 
 print("[startup] app.py fully loaded — regen handlers registered, SSR disabled")
 demo.queue(max_size=10).launch(ssr_mode=False, height=900, allowed_paths=["/tmp"])