feat: add trn2.3xlarge to neuron-device-plugin.yaml

[gabrielzpascual]

What's changing and why?

Add trn2.3xlarge to neuron-device-plugin.yaml to enable trn2.3xlarge support on HyperPod

Before/After UX

Before:
trn2.3xlarge would not have neuron-device-plugin pod scheduled on it.

After:
trn2.3xlarge has neuron-device-plugin pod scheduled on it.

How was this change tested?

Created HyperPod cluster and scaled up trn2.3xlarge instances and verified pods run:

dev-dsk-gabepasc-2b-3b04a487 % kubectl get pods -A
NAMESPACE      NAME                                   READY   STATUS              RESTARTS       AGE
aws-hyperpod   health-monitoring-agent-gk8gd          0/1     RunContainerError   955 (1s ago)   3d9h
kube-system    aws-node-6ktcj                         2/2     Running             0              3d11h
kube-system    coredns-7468d594c9-fsfzh               1/1     Running             0              3d13h
kube-system    coredns-7468d594c9-nd7v9               1/1     Running             0              3d13h
kube-system    eks-pod-identity-agent-px8bb           1/1     Running             0              3d11h
kube-system    kube-proxy-6qgjs                       1/1     Running             0              3d11h
kube-system    neuron-device-plugin-daemonset-hk6zg   1/1     Running             0              31s

dev-dsk-gabepasc-2b-3b04a487 % kubectl describe node hyperpod-i-0f7d6c2b34324bafd | grep -E "neurondevice|efa|cpu"
  aws.amazon.com/neurondevice:  1
  cpu:                          12
  aws.amazon.com/neurondevice:  1
  cpu:                          11900m
  cpu                          850m (7%)   500m (4%)
  aws.amazon.com/neurondevice  0           0

Are unit tests added?

NA

Are integration tests added?

NA

Reviewer Guidelines

‼️ Merge Requirements: PRs with failing integration tests cannot be merged without justification.

One of the following must be true:

• All automated PR checks pass
• Failed tests include local run results/screenshots proving they work
• Changes are documentation-only