Panic just after Session was successfully started

## The Problem

I'm trying to use `session-manager-plugin` to log into my ECS Fargate containers and I face the following issue:

```sh
Starting session with SessionId: ecs-execute-command-*****************
root@ip-***-***-***-***:/src# panic: runtime error: invalid memory address or nil pointer dereference
[signal SIGSEGV: segmentation violation code=0x2 addr=0x18 pc=0x102584cec]

goroutine 1 [running]:
github.com/aws/session-manager-plugin/src/sessionmanagerplugin/session.glob..func2(0x140001a0340, {0x102980dd0, 0x140001d6760})
/Users/kinjal/go/pkg/mod/github.com/aws/session-manager-plugin@v0.0.0-20230315220744-7b544e9f381d/src/sessionmanagerplugin/session/session.go:98 +0x5c
github.com/aws/session-manager-plugin/src/sessionmanagerplugin/session.(*Session).Execute(0x140001a0340, {0x102980dd0, 0x140001d6760})
/Users/kinjal/go/pkg/mod/github.com/aws/session-manager-plugin@v0.0.0-20230315220744-7b544e9f381d/src/sessionmanagerplugin/session/session.go:233 +0x21c
github.com/wingsure-inc/sshing/sshing.EcsExec()
/Users/kinjal/Desktop/DRIVE/WORK/sshing/sshing/aws.go:179 +0xa40
main.main()
/Users/kinjal/Desktop/DRIVE/WORK/sshing/main.go:9 +0x1c
exit status 2
```

> Sensitive information have been masked with ***

From the log line above, I can see that the session was successfully started. The line `root@ip-***-***-***-***:/src#` clearly depicts the same. I also confirmed through the Session Manager in AWS Console (https://ap-south-1.console.aws.amazon.com/systems-manager/session-manager/sessions?region=ap-south-1) that the sessions were started successfully. However as soon as the session is created, it `panics` somehow!

## More information

I've setup the session in the following fashion:

```go
var ssmSession session.Session
ssmSession.SessionId = *output.Session.SessionId
ssmSession.StreamUrl = *output.Session.StreamUrl
ssmSession.TokenValue = *output.Session.TokenValue
ssmSession.Endpoint = endpoint.URL
ssmSession.TargetId = fmt.Sprintf("ecs:%v_%v_%v", clusterName, taskId, containerRuntimeId)
ssmSession.ClientId = uuid.NewV4().String()
ssmSession.DataChannel = &datachannel.DataChannel{}
// ssmSession.SessionType = "NonInteractiveCommands"

log := log.Logger(true, ssmSession.ClientId)
err = ssmSession.Execute(log)
HandleErr(err)
```

The above code is inspired by this:
https://github.com/aws/session-manager-plugin/blob/7b544e9f381d809fd7117747d4b78b244addcf1e/src/sessionmanagerplugin/session/session.go#L182-L209

## AWS CLI

However, when I try to log into the container through `aws-cli` using this command, it works:
```sh
aws ecs execute-command --region --cluster --task --container --command "/bin/bash" --interactive
```

Digging into the `aws-cli` code, I find that it invokes `session-manager-plugin` as a `subprocess` rather than implementing it programmatically:
https://github.com/aws/aws-cli/blob/5218ac17b8141e1a15afed9e644904a6a1081b7d/awscli/customizations/ecs/executecommand.py#L103-L111

## My Goal
I wish to log into my ECS Fargate containers programmatically! My goal is to reverse-engineer `aws-cli` and `session-manager-plugin` so that they work together

Any help would be appreciated. Thanks in advanced!