Fix/preflight NUMA imbalance to mean uneven GPU distribution across nodes (#554)

### Changes:
Only flag imbalance if the COUNT of GPUs on each node differs.
Example: 
4 on Node 0, 4 on Node 1 -> counts=[4,4] -> set={4} -> len=1 -> NOT
imbalanced.
7 on Node 0, 1 on Node 1 -> counts=[7,1] -> set={7,1} -> len=2 ->
Imbalanced.

### Reason for changes:
The previous logic would issue a NUMA imbalance warning if not all GPUs
were connected to the same node, resulting in a false positive when
using a multi-socket CPU.

---------

Co-authored-by: Xiaoming-AMD <Xiaoming.Peng@amd.com>

Author: alexsu52

primus/tools/preflight/gpu/gpu_topology.py

@@ -7,6 +7,7 @@
 from __future__ import annotations
 
 import os
+from collections import Counter
 from typing import Any, Dict, List, Optional
 
 from .gpu_probe import probe_gpus
@@ -109,7 +110,10 @@ def run_gpu_standard_checks(*, force_topology: bool = False) -> Dict[str, Any]:
         findings.append(Finding("warn", "NUMA mapping unavailable (amd-smi not found); skipped", {}))
     else:
         nodes = [x.get("numa_node") for x in numa.get("gpus", []) if x.get("numa_node") is not None]
-        imbalance = len(set(nodes)) > 1 if nodes else False
+        imbalance = False
+        if nodes:
+            counts = Counter(nodes).values()
+            imbalance = len(set(counts)) > 1
         findings.append(
             Finding("info", "GPU↔NUMA mapping", {"mapping": numa.get("gpus", []), "imbalance": imbalance})
         )