perf: batch decode stream callbacks into one dispatch per step

- add BatchedStreamCallback sink type: TaskManager resolves a decode step's (task_id, token) events under one lock and delivers each sink a single list instead of one call per token
- keep the plain Callable[[str]] callback contract: per-token callbacks still receive one call per event, and invoke_callback/cancel_task wrap single events for batched sinks
- collect aborted, text, and finish STOP events in the scheduler decode loop and dispatch once per step instead of once per token
- register one _ResultSink per generate call (replacing per-task closures) so GenerateResult takes its lock and wakes waiters once per step, with late-bind replay for tasks that start decoding before add_task returns their id
- apply GenerateResult batches under a single condition hold via append_batch; append delegates to it
- update engine test fakes to the batched contract and add coverage for event grouping, single-event dispatch, cancel STOP, and late-bind replay

Benchmark: NVIDIA L20 (idle), CUDA 12.8, torch 2.11.0+cu128, 1.2B bf16 checkpoint, prompt 512, 256 greedy tokens, CUDA graph on, serving-level decode, 3 trials
- batch 32: 7.808 -> 7.506 ms/token (4098 -> 4263 batch tok/s, +4.0%)
- batch 1/8: unchanged within noise (3.768 -> 3.797 / 4.699 -> 4.607 ms/token)
- full suite: 896 passed

astrai/inference/scheduler.py

@@ -264,17 +264,19 @@ class InferenceScheduler:
 
                     decoded, aborted = self._stepper.step(active)
 
-                    for t in aborted:
-                        self._task_mgr.invoke_callback(t.task_id, STOP)
-
+                    # One dispatch per step: batch-aware sinks take their
+                    # lock (and wake waiters) once instead of once per token.
+                    events: List[Tuple[str, Any]] = [(t.task_id, STOP) for t in aborted]
                     for t in decoded:
                         if t.status == TaskStatus.ABORTED:
                             continue
                         new_text = t.decode_new_token(self._task_mgr.tokenizer)
                         if new_text:
-                            self._task_mgr.invoke_callback(t.task_id, new_text)
+                            events.append((t.task_id, new_text))
                         if t.is_finished(stop_ids):
-                            self._task_mgr.invoke_callback(t.task_id, STOP)
+                            events.append((t.task_id, STOP))
+                    if events:
+                        self._task_mgr.invoke_callbacks(events)
 
         except Exception as e:
             self._stop_event.set()