Power Lora Loader - add lora_prompt string

[HarlequinD]

Proposal:
Change power_lora_loader.py to include a 3rd output as string, and build a auto1111-esq string of loras + weights which can be used as metadata later in a flow.

Need:
minor amends to RETURN_TYPES and RETURN_NAMES for new string output
Format and concat the lora info during the loop, such as:

 # Format as an Auto1111-style LoRA string: <lora:name:strength>
 lora_name = value['lora'].split('/')[-1].split('.')[0]  # Strip path and extension
 lora_line = f"<lora:{lora_name}:{strength_model:.2f}>"
lora_lines.append(lora_line)
lora_prompt = " ".join(lora_lines)

Quick concept test (not fully QC'ed)

from nodes import LoraLoader
from .constants import get_category, get_name
from .power_prompt_utils import get_lora_by_filename
from .utils import FlexibleOptionalInputType, any_type

class RgthreePowerLoraLoader:
  """ The Power Lora Loader is a powerful, flexible node to add multiple loras to a model/clip."""

  NAME = get_name('Power Lora Loader')
  CATEGORY = get_category()

  @classmethod
  def INPUT_TYPES(cls):  # pylint: disable = invalid-name, missing-function-docstring
    return {
      "required": {
        "model": ("MODEL",),
        "clip": ("CLIP",),
      },
      "optional": FlexibleOptionalInputType(any_type),
      "hidden": {},
    }

  RETURN_TYPES = ("MODEL", "CLIP", "STRING")
  RETURN_NAMES = ("MODEL", "CLIP", "lora_prompt")
  FUNCTION = "load_loras"

  def load_loras(self, model, clip, **kwargs):
    """Loops over the provided loras in kwargs and applies valid ones."""
    lora_lines = []

    for key, value in kwargs.items():
      key = key.upper()
      if key.startswith('LORA_') and 'on' in value and 'lora' in value and 'strength' in value:
        strength_model = value['strength']
        strength_clip = value.get('strengthTwo', strength_model)
        if value['on'] and (strength_model != 0 or strength_clip != 0):
          lora = get_lora_by_filename(value['lora'], log_node=self.NAME)
          if lora is not None:
            model, clip = LoraLoader().load_lora(model, clip, lora, strength_model, strength_clip)

            # Format as an Auto1111-style LoRA string: <lora:name:strength>
            lora_name = value['lora'].split('/')[-1].split('.')[0]  # Strip path and extension
            lora_line = f"<lora:{lora_name}:{strength_model:.2f}>"
            lora_lines.append(lora_line)

    lora_prompt = " ".join(lora_lines)
    return (model, clip, lora_prompt)

Result test:

[HarlequinD]

Seems to work well with further testing. Flow checked:
Power Lora Loader->lora_prompt (outbound) combine with Positive CLIP->TEXT (outbound) using Merge string node, passed through Big Context nodes to Image Saver (https://github.com/farizrifqi/ComfyUI-Image-Saver) ->positive (inbound) results in a native read of loras during civitai upload: