"""Parse a response string formatted with specific XML-like tags into a structured dictionary.

The function expects the input string to follow the format: 
`<think><observation>...</observation><reasoning>...</reasoning><prediction>...</prediction></think><answer>...</answer>`.
It extracts content from the tags and organizes it into a dictionary with specific keys.

Args:
    response (str): The input string containing the XML-like tags.
    special_token_list (List[str], optional): A list of special tokens to remove from the extracted content. Defaults to None.
    action_sep (str, optional): The separator used to split the action content into individual actions. Defaults to ','.
    max_actions (int, optional): The maximum number of actions to extract from the action content. Defaults to 3.

Returns:
    Dict: A dictionary with the following keys:
        - llm_raw_response: The original response string.
        - llm_response: The response string with all tags.
        - observation_content: The content inside the `<observation>` tag.
        - reasoning_content: The content inside the `<reasoning>` tag.
        - prediction_content: The content inside the `<prediction>` tag.
        - think_content: The entire content inside the `<think>` tag.
        - action_content: The content inside the `<answer>` tag.
        - actions: A list of actions extracted from the action content.
        - format_correct: A boolean indicating whether the response strictly follows the expected format.

The function also handles special tokens by removing them from the extracted content and limits the number of actions to `max_actions`.
"""
You should write code starting with:
import re
from typing import Dict, List

def parse_grounding_worldmodeling(response: str, special_token_list: List[str] = None, action_sep: str = ',', max_actions: int = 3) -> Dict:
    # TODO: Implement this function
    pass

