More robust pygrb_page_tables

[pannarale]

Motivation

When recently running a light PyGRB workflow, the results webpage was not displaying the loudest onsource trigger information even though the workflow had found one. Additionally, the follow up of the trigger did not happen.

Contents

There are 2 changes in the PR.

1. The call to load the onsource data in page_tables was asking for all timeslides (slide_id="all"), but the onsource is in zero-lag data so it now uses slide_id=0. This fixes the issue reported in the motivation.
2. Digging through the code for the source of the issue, I spotted an if/elif structure that was heavily relying on lengths of arrays. I changed it in order to make it more robust. I should note that now this relies on the words "search" and "missed" possibily being in the hdf5 keys: if the keys structure were to change, the first if condition could become always false. However, this new structure is clear of surviving because of corner cases in array lengths.

Standard information about the request

This is a: bug fix

This change affects: PyGRB

This change changes: result presentation, scientific output

Testing performed

The run that raised the issue is available here.
The new webpage, with the fixes, is instead here.

• The author of this pull request confirms they will adhere to the code of conduct