Parse lora weights from civitai resources

[anonstash]

Description

• Enables directly parsing the Lora weights from the "Civitai resources:" string that is included in an image's metadata when downloaded from Civitai. You can see this in action by clicking the blue arrow to "Read generation params" or when clicking "Send to txt2img" and such
• Adds logic in extensions-builtin/Lora/networks.py to check for a .json file with the same name as a model's file, and extracts the modelVersionId which is what Civitai uses to identify a specific model + version
• Adds a function to modules/infotext_utils.py to handle parsing the resources string
• I've opened a separate PR to the Civitai browser plus plugin to record the modelVersionId in the model json, but this works regardless assuming you have a json file with the modelVersionId as one of the fields

Example prompt:

score_9, score_8_up, score_7_up, car, tree, hill
Negative prompt: score_6, score_5, score_4, simple background, black and white, mono color
Steps: 30, Sampler: DPM++ 2M Karras, CFG scale: 13.5, Seed: 839351144, Size: 832x1216, Clip skip: 2, Civitai resources: [{"type":"checkpoint","modelVersionId":290640},{"type":"ImageJobNetworkParams { Strength = 0.4, TriggerWord = , Type = lora }","weight":0.4,"modelVersionId":426797},{"type":"ImageJobNetworkParams { Strength = 0.4, TriggerWord = , Type = lora }","weight":0.4,"modelVersionId":135867},{"type":"embed","weight":1,"modelVersionId":94057},{"type":"embed","modelVersionId":250708},{"type":"embed","modelVersionId":250712}]

Final prompt:

score_9, score_8_up, score_7_up, car, tree, hill <lora:43stl1ght1ngXLP2:0.4> <lora:add-detail-xl:0.4>

Screenshots/videos:

demo.mp4

Checklist:

• I have read contributing wiki page
• I have performed a self-review of my own code
• My code follows the style guidelines
• My code passes tests

[w-e-w]

I think this should be implemented in an extension using script_callbacks.on_infotext_pasted