Read NVLink and GPU-to-adapter distance from nvidia-smi topo

The release binary is built without cgo, so without NVML, and NVML was
the only source of NVLink peers: on every NVLink machine the portable
agent reported none, and tensor-parallel groups were chosen by GPU index.
Discovery now reads `nvidia-smi topo -m` after either path, for the
NVLink peers (kept from NVML where it gave them), each card's NUMA node,
and how far each network adapter is from it - PIX, PXB, PHB, NODE or
SYS - which travels to the control plane as nic_paths for placement to
use. Both the NICn-with-legend layout of newer drivers and the
adapter-named columns of older ones are read.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

internal/control/agentsvc/convert.go

@@ -3,6 +3,7 @@ package agentsvc
 
 import (
 	"fmt"
+	"maps"
 	"time"
 
 	llmgwv1 "gitea.mynodes.xyz/ilker/llmgw/api/gen/llmgw/v1"
@@ -59,6 +60,7 @@ func NodeFromRegister(r *llmgwv1.Register) inventory.Node {
 			UtilizationPercent: int(g.GetUtilizationPercent()),
 			CUDAVersion:        g.GetCudaVersion(),
 			NVLinkPeers:        intSlice(g.GetNvlinkPeers()),
+			NICPaths:           maps.Clone(g.GetNicPaths()),
 			Health: discovery.GPUHealth{
 				ECCErrors:    int(g.GetEccErrors()),
 				TemperatureC: int(g.GetTemperatureC()),
@@ -112,6 +114,7 @@ func RegisterFromInventory(nodeID, hostname, mgmtAddr, agentVersion string,
 			UtilizationPercent: int32(g.UtilizationPercent),
 			CudaVersion:        g.CUDAVersion,
 			NvlinkPeers:        int32Slice(g.NVLinkPeers),
+			NicPaths:           maps.Clone(g.NICPaths),
 			Problems:           g.Health.Problems,
 			TemperatureC:       int32(g.Health.TemperatureC),
 			EccErrors:          int32(g.Health.ECCErrors),