PR_DESCRIPTION_1020.md

PR Description: Pagination in AchievementsService (#1020)

Overview

This PR addresses the unbounded reads in AchievementsService. Methods that used to retrieve the entire sets of achievements or user progress now accept pagination parameters and perform bounded queries to avoid loading large lists into memory. Additionally, the achievement definition set is now cached and getUserAchievementOverview performs an aggregated query instead of retrieving arrays into memory to calculate totals.

Changes Made

• Added pagination parameters (query?: PaginationQueryDto) to getAllAchievements, getAchievementsByType, getUserAllProgress, and getUserAchievements.
• Updated these methods to return OffsetPaginatedResponse using buildOffsetResponse, and implemented skip/take logic in TypeORM queries.
• Injected CacheManager to cache total_achievements and invalidated the cache (achievements_definitions) on mutations (createAchievement, updateAchievement, deactivateAchievement).
• Refactored getUserAchievementOverview to replace lines 472-476 (loading all active achievements and all user achievements into arrays). It now uses CacheManager for totalAchievements and a single createQueryBuilder aggregation to calculate unlockedCount, totalPoints, and totalExperience.
• Ensured indexes exist on userId within the progress tracking components for optimization.

Acceptance Criteria

• All achievement list endpoints return bounded pages.
• Definition reads are served from cache between mutations.
• The definitions-plus-progress path issues one query rather than two full reads.
• User-scoped achievement lookups are index-backed.