fix broken progress api and current image compatibility

Update ko_KR.json

New options in scripts

Release processing resources after it finishes

Add mid-kdiffusion cfgdenoiser script callback - access latents, conditionings and sigmas mid-sampling

prompts_from_file: allow random seeds to be preserved for the list of prompts

Update for extensions tab and other minor fizes

First commit of november.
Extensions tab localized and other minor translations fixes.

To other portuguese users: se você tiver sugestões de melhores traduções conforme uso, favor entrar em contato ou enviar PRs referentes a localização.
Também seria bem-vindo lusófonos de outros países, caso queiram discutir uma convergência entre diferentes variações do portguês.

Italian localization - Additions and updates

Update de_DE.json

Update KR translation

More strings in the Extensions tab

prototype interrupt api

Extend extras image cache key with upscale_first arg

Added "--clip-models-path" switch

Allow TI training using 6GB VRAM when xformers is available

Add PNG info to pngs only if option is enabled.

Caused error in latest pytorch-lightning==1.8.0

Added translations for these Extensions/Scripts:

Dynamic Prompts
Alpha Canvas
Artists to study
Aesthetic Score

Added a few missing translations and corrected some others. Updated to the latest Extension management tool version.

While I was able to translate in the statistics the text "Time taken:" because the timing itself is a separate label, I couldn't do the same with the labels "Torch active/reserved: 3881/3892 MiB" and "Sys VRAM: 4859/8192 MiB (59.31%)" because the values (memory, percentage) are embedded in the labels (or perhaps I am not enough acknowledged to be sure I am doing it right simply translating the text).

make save dir if save dir is not exists

Correct before image saved callback

add some more classes to safety module for an extension

Unload sd_model before loading the other to solve the issue #3449