feat(cache): implement database fallback for conversation sequence management

Improve the reliability of conversation sequence generation by implementing a fallback mechanism to the database when Redis is unavailable or when keys are missing.

- Add `GetNextSeq` to `MessageRepository` and implement it in `MessageRepositoryAdapter`.
- Update `ConversationCache` to fallback to DB in `GetConvMaxSeq`, `GetConvMaxSeqBatch`, and `GetNextSeq`.
- Ensure Redis TTL is refreshed after every `INCR` operation to prevent sequence counter expiration.
- Refactor service layers to handle sequence generation errors more explicitly.
- Add `IsEnabled` helper to `LayeredCache` for cleaner availability checks.

internal/cache/layered_cache.go

@@ -352,8 +352,13 @@ func (c *LayeredCache) ZCard(ctx context.Context, key string) (int64, error) {
 	return c.redis.ZCard(ctx, key)
 }
 
+// IsEnabled 返回 Redis 是否可用
+func (c *LayeredCache) IsEnabled() bool {
+	return c.enabled && c.redis != nil
+}
+
 func (c *LayeredCache) Incr(ctx context.Context, key string) (int64, error) {
-	if !c.enabled || c.redis == nil {
+	if !c.IsEnabled() {
 		return 0, ErrKeyNotFound
 	}
 	c.local.Delete(key)