Correct label and biological process issues for model annotation loader, possible other improvements

Looking at the output of a loader run: http://tomodachi.berkeleybop.org/amigo/search/model_annotation

The values in the return documents that drive this interface are like:

      {
        "document_category":"model_annotation",
        "id":"gomodel:581e072c00000062|http://model.geneontology.org/581e072c00000062/581e072c00000098|FB:FBgn0005672|GO:0048019|",
        "annotation_unit":"gomodel:581e072c00000062|http://model.geneontology.org/581e072c00000062/581e072c00000098|FB:FBgn0005672|GO:0048019|",
        "annotation_unit_label":"gomodel:581e072c00000062 spi Dmel GO:0048019",
        "annotation_unit_label_searchable":"gomodel:581e072c00000062 spi Dmel GO:0048019",
        "model":"gomodel:581e072c00000062",
        "model_label":"EGFR_SPI_DMEL",
        "model_label_searchable":"EGFR_SPI_DMEL",
        "model_label_searchable":"EGFR_SPI_DMEL",
        "model_url":"http://noctua.berkeleybop.org/editor/graph/581e072c00000062.ttl",
        "model_state":"development",
        "model_date":"2016-11-07",
        "model_date_searchable":"2016-11-07",
        "model_date_searchable":"2016-11-07",
        "enabled_by":"FB:FBgn0005672",
        "enabled_by_searchable":"FB:FBgn0005672",
        "enabled_by_label":"spi Dmel",
        "enabled_by_label_searchable":"spi Dmel",
        "enabled_by_label_searchable":"spi Dmel",
        "function_class":"GO:0048019",
        "function_class_label":"GO:0048019",
        "function_class_label_searchable":"GO:0048019",
        "function_class_label_searchable":"GO:0048019",
        "function_class_closure_map":"{GO:0048019=null}",
        "location_list_closure_map":"{GO:0005615=null}",
        "owl_blob_json":"JSON_BLOB_REMOVED_FOR_BREVITY",
        "evidence_type":"ECO:0000315",
        "evidence_type_label":"ECO:0000315",
        "evidence_type_label_searchable":"ECO:0000315",
        "evidence_type_label_searchable":"ECO:0000315",
        "evidence_type_closure_map":"{ECO:0000315=null}",
        "reference":["PMID:7833286","PMID:7833286\t"],
        "contributor":["http://orcid.org/0000-0003-3212-6364"],
        "evidence_type_closure":["ECO:0000315"],
        "reference_searchable":[
          "PMID:7833286",
          "PMID:7833286\t",
          "PMID:7833286",
          "PMID:7833286\t"],
        "contributor_searchable":[
          "http://orcid.org/0000-0003-3212-6364",
          "http://orcid.org/0000-0003-3212-6364"],
        "location_list_closure":["GO:0005615"],
        "function_class_closure":["GO:0048019"],
        "location_list":["GO:0005615"],
        "location_list_label":["GO:0005615"],
        "score":1.0
}

There are several oddities in this load:

• The id and annotation_unit are generated as a pipe-separated combination of model id, M, P, and C; it seems that one of those is missing as there is a trailing pipe
• all *_searchable fields seem to be repeats; this may not be a "real" issue, practically speaking
• reference and reference_searchable seem to contain tabs; they should not
• all *_closure_map fields contain a "null", which would seem to indicate some issue with JSON deserialization or similar; the form should be id:label; note that this could merely be that there is something wrong with labels (see below)
• for *_label*, in almost all cases, there are no labels, just ids; it could be that the ontology information to get label is not loading properly as part of the command or something deeper
• there should be a field set process_class, just as there is function_class; no process fields exist

There are also a couple of issues that should be coordinated with other development, such as either switching to names for contributor (or making an id/label map), adding groups, etc.

While these may all be separate tickets, the tooling to get at the issues is similar for starters.