[Bug]: taints explanations incorrect in Graviton section

### Installation method

Own AWS account

### What happened?

in this file -
https://github.com/aws-samples/eks-workshop-v2/blob/main/website/docs/fundamentals/compute/managed-node-groups/graviton/scheduling-graviton.md

It says "We have pre-configured our tainted managed node group with a label of tainted=yes that we can use with a nodeSelector."
but then it uses the nodeselector with a different label than "tainted=yes":
```
nodeSelector:
kubernetes.io/arch: arm64
```

Also, then it says "We can see that we have pinned the pod to any node with the tainted=yes label", but earlier, as we saw, we pinned it to arm64, not tainted=yes.

So basically, need to change all tainted=yes labels in the explanations to kubernetes.io/arch=arm64.

### What did you expect to happen?

from the explanations? to see tainted=yes labels/node-selectors, but actually "kubernetes.io/arch=arm64" was used all the way.

### How can we reproduce it?

just follow the current instructions and you'll see we don't have tainted=yes anywhere, but kubernetes.io/arch=arm64, thus need to modify the explanations.

### Anything else we need to know?

_No response_

### EKS version

not relevant

Open website/docs/fundamentals/compute/managed-node-groups/graviton/scheduling-graviton.md and compare the taints explanations with the nodeSelector example. Replace the inconsistent tainted=yes references so the prose matches kubernetes.io/arch=arm64 throughout. Done means the instructions and explanations consistently describe the labels used by the example.

Written by the indexing model from the issue text.