feat: add diseasesWithParents to allele search results (SCRUM-5829)

[oblodgett]

Summary

• Add diseasesWithParents field to AlleleSearchResultDocument and copy it from AlleleSummaryDocument in the converter
• Reorder AlleleSummaryCurationIndexer to convert derived documents first, then strip transport-only fields via removeTransportFields() before indexing summary documents to ES

Context

Companion to agr_curation release/v0.47.12. The curation side populates diseasesWithParents on the summary document; this change ensures it flows through to the AlleleSearchResultDocument in ES, enabling disease→allele related data links in search results.

Test plan

• Deploy to stage
• Reindex after curation release deploys
• Verify allele search result documents in ES contain diseasesWithParents
• Verify disease search results show Allele related data links

[claude]

Code Review

Changes reviewed: 3 files — adds diseasesWithParents field to allele search results and reorders indexing to convert derived documents before stripping transport-only fields.

Verdict: Looks good. ✅

The changes are clean and follow existing patterns:

1. Indexer reordering (AlleleSummaryCurationIndexer.java): The new order (convert → strip → index) correctly ensures derived documents consume transport fields before they're removed. This is a necessary fix — without it, diseasesWithParents (and potentially other transport-only fields) would be stripped before the converters could read them.

2. Converter (AlleleSearchResultConverter.java): The new diseasesWithParents copy follows the identical null/empty guard pattern used for diseasesAgrSlim right above it.

3. Document model (AlleleSearchResultDocument.java): Simple field addition with Lombok handling getters/setters.

No correctness issues, no performance concerns, no security problems found.