(kt-kernel): add numa_nodes parameter for explicit NUMA node mapping (#1891)

Add numa_nodes parameter to BaseMoEWrapper and all subclasses, allowing
users to explicitly specify which NUMA node IDs to use for subpool
mapping instead of always defaulting to sequential [0, 1, ..., N-1].

This enables running multiple KTransformers instances on different NUMA
nodes of the same machine, e.g. --kt-threadpool-count 1 --kt-numa-nodes 1
to bind to NUMA node 1. Previously this required external numactl
workarounds since subpool_numa_map was hardcoded to start from 0.

kt-kernel/python/utils/moe_kernel.py

@@ -1,7 +1,7 @@
 import os
 import torch
 import ctypes
-from typing import Optional
+from typing import List, Optional
 
 # Use relative imports for package structure
 from ..experts_base import BaseMoEWrapper
@@ -97,6 +97,7 @@ class GeneralMoEWrapper(BaseMoEWrapper):
             cpu_save=cpu_save,
             max_deferred_experts_per_token=max_deferred_experts_per_token,
             method=method,
+            numa_nodes=numa_nodes,
         )
 
         # moe-specific: Check if we should load merged safetensor weights