Fix Torchscript error in latent diffusion models unet network

[Warvito]

Try to include torchscript tests for the latent diffusion unet networks. For this, it might be necessary to remove the use of TimestepEmbedSequential and TimestepBlock by creating blocks exclusive to when we have the time embedding included.
A similar solution to Huggingface diffuser might be useful:
https://github.com/huggingface/diffusers/blob/2d35f6733a2d698e8917896071444a5923993ae7/src/diffusers/models/unet_blocks.py#L461
https://github.com/huggingface/diffusers/blob/2d35f6733a2d698e8917896071444a5923993ae7/src/diffusers/models/unet_blocks.py#L379
https://github.com/huggingface/diffusers/blob/2d35f6733a2d698e8917896071444a5923993ae7/src/diffusers/models/unet_blocks.py#L576

[Warvito]

As highlighted by @SANCHES-Pedro #14 (comment) , it will be better to follow the Huggingface solution, and it will also solve this issue with torchscript. We should adopt this for Version 0.1