perf: migrate all user/session/auth IDs from VARCHAR(36) to native UUID

- Schema: all ~15 VARCHAR(36) columns → UUID with DEFAULT gen_random_uuid()
- Domain entities: User, Session, DeviceCode, AppPassword, Share → id: Uuid
- DTOs: CurrentUser.id → Uuid (API boundary DTOs keep String for JSON)
- Auth middleware: parse JWT claims.sub (String) → Uuid at boundary
- All repository traits, port traits, service impls updated end-to-end
- Handlers: pass Uuid by value (Copy, 16 bytes) instead of String refs
- Settings chain: updated_by column → Uuid (was text, caused setup crash)
- Removed ~650 lines of String↔Uuid conversion boilerplate
- Eliminates per-request heap allocations for ID cloning
- 16-byte binary comparison vs 36-byte string comparison in all queries
- Native UUID indexing in PostgreSQL (btree on 16 bytes vs 36-char text)

85 files changed, 1090 insertions(+), 1739 deletions(-)

src/application/dtos/user_dto.rs

@@ -1,6 +1,7 @@
 use crate::domain::entities::user::User;
 use chrono::{DateTime, Utc};
 use serde::{Deserialize, Serialize};
+use uuid::Uuid;
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct UserDto {
@@ -80,7 +81,7 @@ pub struct RefreshTokenDto {
 /// Authenticated current user data (for use in application services)
 #[derive(Clone, Debug, Serialize, Deserialize)]
 pub struct CurrentUser {
-    pub id: String,
+    pub id: Uuid,
     pub username: String,
     pub email: String,
     pub role: String,