Description of problem:
If user chooses a very small task batch size (e.g. 1 or 2) it may happen that some RunHostJob(s) get stuck in running/pending (suspended) status forever.

One has to then clean/cancel the tasks to get rid of them:

# sudo su - postgres -c "psql -d foreman -c 'select label,count(label),state,result from foreman_tasks_tasks where state <> '\''stopped'\'' group by label,state,result ORDER BY label;'"
                        label                         | count |   state   | result  
------------------------------------------------------+-------+-----------+---------
 Actions::RemoteExecution::RunHostJob                 |     2 | running   | pending
 Actions::RemoteExecution::RunHostsJob                |     1 | running   | pending


Version-Release number of selected component (if applicable):
Satellite 6.12

How reproducible:
Sort of (not 100% deterministic)

Steps to Reproduce:
1. Set batch size to 1:
- Administer -> Settings -> Tasks -> Capsule tasks batch size -> 1
- Administer -> Settings -> Ansible -> Capsule tasks batch size for Ansible -> 1

2. Have a set of hosts (I used 13) and schedule a recurring job for them:
- Monitor -> Jobs -> Run Job -> Ansible Playbook; Ansible Roles - Ansible Default
- Custom cron so the job runs every minute (*/1 * * * *) 
- Repeat N times: 10


Actual results:
- Some job iterations result in forever running RunHostJob(s) tasks. 
- Inspecting the job task shows the following error:
~~~
Error loading data from Capsule: RestClient::NotFound - 404 Not Found
~~~
- For more clarity see attached screenshots.

Expected results:
- The 404 errors do not occur.
- Jobs are not stuck in running/pending state forever.