SIP patron part level holds respond blank

This fixed it for us:

http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/blake/LP1525394_SIP_patron_part_level_holds_respond_blank

While updating __hold_to_title to support part holds is wonderful, I think we also need to pay attention to these functions:

find_copy_for_hold
find_hold_from_copy

As well as include Issuance holds (in all three functions).

I've tested and signed-off on Blake's parts hold addition and have also added issuance holds in this branch: https://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/jboyer/lp1525394_more_sip_hold_titles / working/user/jboyer/lp1525394_more_sip_hold_titles

This should complete our current hold type coverage for this feature.

To test, apply the top 2 commits from that branch, place multiple types of holds for a user (at a minimum for this test, Parts and Issuance) and then send a Patron Information Message (63) from a SIP client or telnet and ask for hold details (set the Summary field first position to 'Y'), and you should see an AS field entry for each hold placed. For example, in the concerto data, an I hold on the IEEE title, T hold on The Concerto, and a P hold on Le Concerto gives this Patron Information Response:

64 Y 00020190909 152100000300000000000000000003AOlocal|AAsip-client|AESIP User|BHUSD|BDnaw nope, mnn 54321|BF555-123-4321|AQBR1|ASIEEE International conference on software testing, verification and validation : [proceedings]|ASThe concerto|ASLe concerto|BLY|CQY|PA20220906|PCSIP|PIFiltered|XI238|

I can affirm that Volume/Issuance holds seem to work.

However, in a Part level hold, without a targeted copy, SIP responds with an incorrect barcode pointing to a completely unrelated item. My part level hold for a DVD points to a John Denver CD. Completely different monograph parts and unrelated items.

There is an additional commit that I neglected to upload to the initial branch. Here's a rebased version that contains all 3: https://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/jboyer/lp1525394_more_sip_hold_titles_2 / working/user/jboyer/lp1525394_more_sip_hold_titles_2

I'd like some more eyes on it and some additional testing if possible. My by-hand testing looked ok with msg64_hold_items_available set to barcode or not, but I'm not using anything fancier than telnet here.