Reorganise "Save jobs" page

[frankharkins]

This page originally submitted a simple job using the cloud simulator as a demonstration. Since the simulators are being deprecated, #1160 removed the job-submitting step and hardcoded the job_id into the notebook. This was a bad idea because it means the notebook breaks if not run with my IBM Quantum account.

This PR reorganises the page to get job_id programatically first. This lets us avoid hardcoding the job_id in the notebook, which means we can run the notebook with any IBM account that has submitted at least one job.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[Eric-Arellano]

Looks good! Are we confident that CI will always have jobs run in the past day, the next time we go to regenerate this notebook? Would it instead be safer to do something like past 3 days or past week? It won't fail if no jobs were run in the past day, but the output would be empty

[frankharkins]

Are we confident that CI will always have jobs run in the past day...

It think it's likely it might not, but IMO empty output is OK. Open to other suggestions though.

[Eric-Arellano]

Open to other suggestions though.

You could change it to the past week. Or probably even better: ask for all jobs but use a slice to only show up to 3, jobs[:3]

>>> [1,2,3,4,5][:3]
[1, 2, 3]
>>> [1][:3]
[1]

[frankharkins]

I quite like the datetime example since it's something a user would realistically want to do, and a bit harder to work out on your own. Maybe we could extend the time period to a year but only display three of them?

[frankharkins]

As a compromise, I went with the last 90 days as timedelta doesn't seem to accept years and 3 months should be more than enough if we're running the cron job every two weeks.