[Feature] Implement Geospatial Indexing for Matching Queries #7

Problem

Nearby donor queries scan entire collection without indexes. Performance degrades as donor database grows.

Impact

• Matching queries become slow
• Cannot scale to thousands of donors
• Geographic queries inefficient

Solution

Add MongoDB 2dsphere indexes on location fields:

• Donor.location (GeoJSON Point)
• Inventory.location (GeoJSON Point)
• Hospital.location (GeoJSON Point)

Acceptance Criteria

• 2dsphere indexes created on all location fields
• Query performance benchmarked
• Nearby queries use indexes (verify with explain)
• Tests confirm performance improvement