Possible memory leak

### Describe the bug

I notice an unexpected memory increase when pooler running and not processing messages. Here the command launched with foreman:
`default_worker: bundle exec aws_active_job_sqs -t 1 --shutdown_timeout 10 --queue default`
`document_worker: bundle exec aws_active_job_sqs -t 1 --shutdown_timeout 10 --queue documents`

and hear a screenshot of memory monitor:


### Expected Behavior

Memory not increase when pooler run and not processing any messages

### Current Behavior

Memory increase when pooler run and not processing messages

### Reproduction Steps

Launch the pooler and monitor the memory usgae.

### Possible Solution

_No response_

### Additional Information/Context

_No response_

### Gem name ('aws-sdk', 'aws-sdk-resources' or service gems like 'aws-sdk-s3') and its version

aws-activejob-sqs (1.0.1) aws-sdk-rails (5.1.0) aws-sdk-sqs (1.93.0)

### Environment details (Version of Ruby, OS environment)

Ruby 3.4.2, Debian GNU/Linux 12 (bookworm)

Start by reproducing the reported idle-memory increase with the two foreman commands in the issue, using Ruby 3.4.2 on Debian 12 and the listed gem versions. Monitor memory while the SQS pooler processes no messages, then inspect the worker and pooler entry points implicated by the command; done means memory remains stable during idle operation. No source file or test is identified in the issue.

Written by the indexing model from the issue text.