fix: add gr.Request to xregen_* for proper ZeroGPU quota attribution

Gradio automatically injects gr.Request into the server-side handler
when the function signature includes it. This gives @spaces.GPU access
to the user's auth token via the request context, so ZeroGPU attributes
GPU usage to the logged-in Pro account instead of the anonymous pool.

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

app.py

@@ -1358,7 +1358,8 @@ def _resample_to_slot_sr(wav: np.ndarray, src_sr: int, dst_sr: int,
 
 def xregen_taro(seg_idx, state_json, slot_id,
                 seed_val, cfg_scale, num_steps, mode,
-                crossfade_s, crossfade_db):
+                crossfade_s, crossfade_db,
+                request: gr.Request = None):
     """Cross-model regen: run TARO inference and splice into *slot_id*."""
     meta    = json.loads(state_json)
     seg_idx = int(seg_idx)
@@ -1381,7 +1382,8 @@ def xregen_taro(seg_idx, state_json, slot_id,
 
 def xregen_mmaudio(seg_idx, state_json, slot_id,
                    prompt, negative_prompt, seed_val,
-                   cfg_strength, num_steps, crossfade_s, crossfade_db):
+                   cfg_strength, num_steps, crossfade_s, crossfade_db,
+                   request: gr.Request = None):
     """Cross-model regen: run MMAudio inference and splice into *slot_id*."""
     meta    = json.loads(state_json)
     seg_idx = int(seg_idx)
@@ -1416,7 +1418,8 @@ def xregen_mmaudio(seg_idx, state_json, slot_id,
 def xregen_hunyuan(seg_idx, state_json, slot_id,
                    prompt, negative_prompt, seed_val,
                    guidance_scale, num_steps, model_size,
-                   crossfade_s, crossfade_db):
+                   crossfade_s, crossfade_db,
+                   request: gr.Request = None):
     """Cross-model regen: run HunyuanFoley inference and splice into *slot_id*."""
     meta    = json.loads(state_json)
     seg_idx = int(seg_idx)