DELPHI logbook title uniqueness and search discoverability

[tiborsimko]

Current behaviour

Two observations regarding DELPHI logbooks after merging and deploying PR #3728:

(1) All record titles are the same, which makes it hard to distinguish different search results visually:

It would be better to always use different titles for different records, as we do for JADE notebooks:

or for JADE computing notes:

(2) There are 11 logbooks, but when one searches for "logbook" one finds only one record:

This could be solved by either correcting the metadata (because currently only one record has the word "logbook" inside the text, whilst the others use "logbooks" in plural), but also by better search engine stemming settings (because when you have a data value like "logbooks", then it could be indexed as a "logbook" with proper stemming, so that it can be retrieved irrespective of whether the user types singular or plural form of the word, as is being usually the case with general-purpose seach engines).

Expected behaviour

The different data records released on the portal should have distinguishable titles and not reuse the same title for different things (as for JADE examples above).

The search engine should better support plural vs singular word matching when doing queries.

CC @psaiz @schwicke

[psaiz]

Thanks for pointing this out. For the first point, #3766 should take care of it.
Regarding the second issue, I would say that the main issue is not the plural/singular, but the uppercarse/lowercase. "Logbook" returns also the 11 entries. The issue is that the type.secondary of the record is stored as keyword (which is needed for the facets). If we store it as well as text, it will be case insensitive, and it will provide the search results defined in the behaviour. This PR should take care of it