perf: move decode split partials to InferenceWorkspace

- Replace per-.cu-file static cached tensors with workspace-managed pre-allocated buffers

- InferenceWorkspace now owns decode_o_part / decode_ml_part (mirrors FlashInfer's workspace pattern)

- KVCache carries the buffers through the backend -> C++ kernel chain

- C++ kernels accept optional pre-allocated buffers; fallback to alloc_split_partials for backward compat

- Pre-allocates once at Executor init, zero allocation in the decode hot loop

- Prerequisite for CUDA-graph capture (all kernel addresses are stable)

tests/inference/test_cache.py

@@ -16,7 +16,12 @@ from astrai.inference.core.workspace import InferenceWorkspace
 def _ws(pool: PagePool) -> InferenceWorkspace:
     """Workspace sized to the pool (bind_tasks requires it)."""
     return InferenceWorkspace(
-        pool.max_batch_size, pool.max_seq_len, pool.device, pool.dtype
+        pool.max_batch_size,
+        pool.max_seq_len,
+        max_q_heads=2,
+        head_dim=4,
+        device=pool.device,
+        dtype=pool.dtype,
     )