fix(distributed-demo): make apply_chat_template token-id coercion transformers 4.x AND 5.x safe

#105's demo assumed apply_chat_template returns token ids (transformers 4.x);
on transformers 5.x it returns a string -> verifier.prefill got a str and raised
'str object cannot be interpreted as an integer' (hit on the Mac, whose venv runs
transformers 5.x for gemma-4). Pass tokenize=True and coerce str/BatchEncoding/
nested-list to a flat List[int]. Verified on the GPU host (transformers 4.57) and
re-validated on the Mac (5.x).

Co-authored-by: FluffyAIcode <FluffyAIcode@users.noreply.github.com>

Author: cursoragent

scripts/demo_distributed_spec_decode.py

@@ -171,14 +171,27 @@ async def _run_verifier_node(args: argparse.Namespace) -> int:
         dtype=torch.bfloat16, device="cpu",
         sink_size=args.sink, window_size=args.window,
     ))
-    prompt_ids = verifier.tokenizer.apply_chat_template(
+    # Echo-style answers are where the n-gram proposer shines; Qwen3's thinking
+    # preamble is novel text the lookup cannot draft. Templates without the
+    # variable ignore it harmlessly.
+    # transformers 4.x returns token ids from apply_chat_template(tokenize=True);
+    # 5.x can return a string (or a BatchEncoding) — coerce to a flat List[int]
+    # so this works across both (the Mac engine needs transformers 5.x).
+    _templated = verifier.tokenizer.apply_chat_template(
         [{"role": "user", "content": args.prompt}],
-        add_generation_prompt=True,
-        # Echo-style answers are where the n-gram proposer shines;
-        # Qwen3's thinking preamble is novel text the lookup cannot
-        # draft. Templates without the variable ignore it harmlessly.
+        add_generation_prompt=True, tokenize=True,
         enable_thinking=args.enable_thinking,
     )
+    if isinstance(_templated, str):
+        prompt_ids = list(verifier.tokenizer.encode(_templated))
+    else:
+        if hasattr(_templated, "input_ids"):
+            _templated = _templated.input_ids
+        if hasattr(_templated, "tolist"):
+            _templated = _templated.tolist()
+        if _templated and isinstance(_templated[0], (list, tuple)):
+            _templated = _templated[0]
+        prompt_ids = [int(x) for x in _templated]
 
     # --- 4. Greedy baseline (same verifier, local only) --------------
     t0 = time.perf_counter()