Upgrade EBS volume type from gp2 to gp3 in EC2 launch templates

[igor-soldev]

What does this PR do?

This PR updates the default EBS volume type from gp2 to gp3 in the EC2 launch templates for both terraform-ecs-ec2 and terraform-ecs-ec2-sqs deployment setups.

Why is this change necessary?

As discussed with Carter on the Sequin Community Slack, upgrading to gp3 provides immediate benefits for anyone self-hosting Sequin on AWS:

1. Cost Reduction: gp3 volumes are up to 20% cheaper per GB compared to gp2.
2. Performance: gp3 provides a baseline of 3,000 IOPS and 125 MB/s throughput regardless of volume size, eliminating the burst-credit limitations inherent to gp2.

Additionally, I removed the iops = "0" parameter, as gp3 volumes require either a minimum of 3000 IOPS or the parameter can be omitted entirely (AWS defaults it to 3000 for gp3). Setting iops = 0 with gp3 will cause a Terraform validation error.

(Note: The diff shows iops = "0" being removed, which is the correct approach when migrating to gp3 without explicitly defining custom IOPS).

How was this identified?

This opportunity was identified while running a static infrastructure analysis on the deployment/ directory using InfraScan (an open-source FinOps/SecOps tool by SolDevelo).

You can view the full report that highlighted this here:
View InfraScan Report for Sequin

Testing

• Validated Terraform syntax.
• Confirmed AWS provider documentation supports implicit IOPS for gp3.