[aws-cloudwatch-metrics]: node_filesystem_utilization is not working correctly

**Describe the bug**
Installed this chart in my EKS and found that the value of the metric, **node_filesystem_utilization**, is not same as checked on the ec2 instance in EKS cluster. There is a bid difference between 2 values.

What does the metric, node_filesystem_utilization, collect?

Btw, I install cloudwatch agent manually on my standalone EC2 instances and see the metric, **disk_used_percent**. The value of this metric is same as the value checked from the host. After this installation, I see the CWAgeng namespace. What I see is the "ContainerInsights" after the installation of this chart.

I have a very simple setting for aws-cloudwatch-metrics
```
clusterName: "my-eks-cluster"
serviceAccount:
name: "my-eks-cluster-cloudwatch-agent-sa"

```

**Steps to reproduce**
1: install this chart into the EKS cluster

**Expected outcome**
Expect to collect the right metrics with right values.

**Environment**

* Chart name: aws-cloudwatch-metrics
* Chart version: latest
* Kubernetes version: 1.24
* Using EKS (yes/no), if so version? yes

Start by reproducing the aws-cloudwatch-metrics chart installation on EKS using the minimal values shown. Inspect the chart's CloudWatch agent configuration and compare node_filesystem_utilization with the host filesystem value and disk_used_percent. Done means identifying the metric's collection semantics and correcting or clearly documenting the discrepancy.

Written by the indexing model from the issue text.