Bug 36649: Correctly embed biblio when retrieving items when adding to a train

Bug 36649: Correctly embed biblio when retrieving items when adding to a train

I didn't find when this regression has been introduced but the third
parameter here is the headers, not parameters, we should not repeat
"headers".

Test plan:
1. Add a new processing that contains columns from the database, for example biblio.title and biblio.author.
2. Create a new train.
3. Add items to the waiting list.
4. Click on "Add last x items to the train" button.
5. Choose the train you created on number 2.
6. Choose the processing you added on number 1.
7. Try to click the Submit button.
=> Without the patch nothing happens and there is a JS error in the
console: Uncaught TypeError: item.biblio is undefined
=> With this patch applied the item is correctly added to the train and
the attribute is properly populated.

Signed-off-by: Anneli Österman <anneli.osterman@koha-suomi.fi>
Signed-off-by: Pedro Amorim <pedro.amorim@ptfs-europe.com>
Signed-off-by: Katrin Fischer <katrin.fischer@bsz-bw.de>
(cherry picked from commit 63612311e32a546a6ea1067e0c65ac3199f2f292)
Signed-off-by: Fridolin Somers <fridolin.somers@biblibre.com>