fix: add out_buf to attn_paged_decode for CUDA graph capture compatibility

- Pre-allocate decode_out in InferenceWorkspace so attn_paged_decode does not call torch::empty inside graph capture
- Wire decode_out through KVCache, PagePool.bind_tasks, and CudaBackend.fwd_decode
- Run live forward before graph capture to get valid output (graph pool memory is zeroed after capture block exits)
- Greedy generation with graph replay is bit-exact across all batch sizes
- Decode speedups vs no-graph: B=1 2.09x, B=4 1.80x, B=8 1.94x, B=16 1.76x

astrai/inference/core/graph.py

@@ -86,12 +86,14 @@ class CudaGraphContext:
         if key in self._graphs:
             self._graphs[key].replay()
         elif key in self._warmed:
+            cap_output = model(**kwargs)
             torch.cuda.synchronize()
             graph = torch.cuda.CUDAGraph()
             with torch.cuda.graph(graph):
                 self._outputs[key] = model(**kwargs)
             self._graphs[key] = graph
             self._warmed.discard(key)
+            return cap_output
         else:
             self._warmed.add(key)
             self._outputs[key] = model(**kwargs)