This will be loaded via Stimulus when needed in a subsequent commit.

Alma SRU controller tests

This still needs to be filtered, most likely (tho the AlmaSru model
will also do its own filtering)

This involves adding lazy-loading support to the content loader
controller, which may be a feature that other async calls would
benefit from.

- Also adds additional filter on UI invoking this path, leveraging new validation method

** Why are these changes being introduced:

Rather than render multiple availability statements, we want to just
show the first, with a suffix phrase of "and other locations" if there
are more locations.

** Relevant ticket(s):

* https://mitlibraries.atlassian.net/browse/use-599

** How does this address that need:

This slightly refactors the formatting logic for handling the statements
after they've been formatted. All blocks are formatted to start with,
and then if there are multiple statements we append the suffix. Only the
first statement is ever returned, as a list of one entry - so the
overall contract of the .lookup() method is unchanged.

** Document any side effects to this change:

It is a bit inefficient to format records that we then discard. I'm
not sure that this will be a significant issue, but it should be noted.

** Why are these changes being introduced:

Up until now, the availability statements have been empty links - but
we want them to go to the full records view, and specifically to the
locations block of that page.

** Relevant ticket(s):

* https://mitlibraries.atlassian.net/browse/use-599

** How does this address that need:

This calls the new PrimoLinkBuilder's full_record_link method, and adds
the `getit_link1_0` document fragment to send the user to the list of
available locations.

** Document any side effects to this change:

There are two concerns to note here:
1. The document fragment we are using feels fragile, and like it may
go away in the future. I also haven't checked what the fragment is for
NDE.
2. This link is _very_ similar to the full record link that appears
immediately before this in the UI. It will likely pass an accessibility
check because it is not identical - but this feels like a technicality,
and not an ideal user experience. I'm choosing to propose this anyway to
see if I'm alone in thinking this - I can live with it, but it would not
take much for me to support an alternative.

** Why are these changes being introduced:

Our previous implementation of the availability statements had some nice
features like icons and bold styling for location names. We want to
carry this forward, but want to keep the model-level implementation for
this formatting.

** Relevant ticket(s):

* https://mitlibraries.atlassian.net/browse/use-599

** How does this address that need:

This removes the availability() and location() record helpers, and
reintroduces those features within the AlmaSru model. This includes the
status handling logic of the previous implementation.

** Document any side effects to this change:

Other parts of the application rely on the icon() helper in the record
helper model, so we can't remove that here. As a result, there is now a
duplication of that logic in two places in the application, which isn't
ideal.

Avoiding this would mean changing where the AlmaSru model formats its
output, which I'd rather not do at this point - but I'm open to feedback
about this during code review.

Update test and teardown for AlmaSru.enabled?

Update result_get? check to include Alma ID format

We've removed the availability, so that cause and test were
meaningless.