feat: migrate to DragonflyDB and clean up environment configuration

- Replace Redis with DragonflyDB for better performance and memory efficiency
- Remove redundant environment variables (POSTGRES_*, ENCRYPTION_KEY, OAUTH_SERVICE_URL)
- Consolidate database configuration to use single DB_* variables
- Use JWT_SECRET for both JWT tokens and encryption
- Remove PORT variable redundancy, use BACKEND_PORT consistently
- Clean up docker-compose configurations for dev/prod consistency
- Add DragonflyDB configuration with optimized memory usage
- Remove redis.conf as it's no longer needed
- Update health checks to use Redis-compatible CLI for DragonflyDB

dragonfly.conf

@@ -0,0 +1,112 @@
+# DragonflyDB Configuration for Trackeep
+# 
+# DragonflyDB is a modern Redis-compatible in-memory database
+# Optimized for performance and lower memory usage
+
+# =============================================================================
+# NETWORK
+# =============================================================================
+
+# Accept connections on all interfaces (safe when behind Docker network)
+bind 0.0.0.0
+
+# Default port (same as Redis for compatibility)
+port 6379
+
+# TCP listen() backlog
+tcp-backlog 511
+
+# Close connection after N seconds of idle time (0 = disabled)
+timeout 0
+
+# TCP keepalive
+tcp-keepalive 300
+
+# =============================================================================
+# SECURITY
+# =============================================================================
+
+# Require password for connections
+# Set via environment variable: requirepass ${DRAGONFLY_PASSWORD}
+requirepass dragonfly123
+
+# Disable dangerous commands in production
+rename-command FLUSHDB ""
+rename-command FLUSHALL ""
+rename-command CONFIG "CONFIG_9f8a2b3c"
+rename-command DEBUG ""
+rename-command SHUTDOWN "SHUTDOWN_7d4e1f9a"
+
+# =============================================================================
+# MEMORY MANAGEMENT
+# =============================================================================
+
+# Maximum memory limit (256MB suitable for small-medium deployments)
+# DragonflyDB is more memory efficient than Redis
+maxmemory 256mb
+
+# Eviction policy when maxmemory is reached
+# allkeys-lru: Remove less recently used keys first (recommended for caching)
+maxmemory-policy allkeys-lru
+
+# =============================================================================
+# PERSISTENCE
+# =============================================================================
+
+# Enable AOF persistence (recommended for session durability)
+appendonly yes
+
+# AOF file name
+appendfilename "appendonly.aof"
+
+# Sync strategy: everysec (recommended balance)
+appendfsync everysec
+
+# Auto-rewrite AOF when it grows by X%
+auto-aof-rewrite-percentage 100
+
+# Minimum size before auto-rewrite
+auto-aof-rewrite-min-size 64mb
+
+# Working directory for persistence
+dir /data
+
+# =============================================================================
+# CLIENTS & PERFORMANCE
+# =============================================================================
+
+# Maximum number of client connections
+maxclients 10000
+
+# Number of databases (default 16)
+databases 16
+
+# Latency monitoring
+latency-monitor-threshold 100
+
+# Slow log (log queries taking > N microseconds)
+slowlog-log-slower-than 10000
+
+# Slow log max length
+slowlog-max-len 128
+
+# =============================================================================
+# LOGGING
+# =============================================================================
+
+# Log level: debug, verbose, notice, warning
+loglevel notice
+
+# Log file (empty = stdout, good for Docker)
+logfile ""
+
+# =============================================================================
+# DRAGONFLYDB SPECIFIC OPTIMIZATIONS
+# =============================================================================
+
+# Enable DragonflyDB-specific optimizations
+# These are automatically enabled in DragonflyDB
+
+# Better memory management
+# Improved multi-core utilization
+# Enhanced performance for caching workloads