Added plot_metrics multiple comparison Function

[Archillesjakins]

This new function allows users to compare performance metrics and hyper parameters for multiple machine learning models in a single plot. It automatically parses JSON outputs, excludes irrelevant keys (e.g., cache_size, random_state), and visualizes only relevant values, making it easier for ML engineers to compare models.

Key Features:
Supports multiple models comparison.
Automatically parses JSON strings into Python dictionaries.
Excludes specific keys like cache_size, random_state, etc.
Provides customizable and clear visualization for metrics like accuracy, precision, recall, etc.

Usage Example:

Jupiter notebook: https://notebooks.githubusercontent.com/view/ipynb?browser=safari&bypass_fastly=true&color_mode=auto&commit=20e57b0e257bcbd3dc158511dac6f523be332da2&device=unknown_device&docs_host=https%3A%2F%2Fdocs.github.com&enc_url=68747470733a2f2f7261772e67697468756275736572636f6e74656e742e636f6d2f417263686c79323032322f6c6f676c6c6d2f323065353762306532353762636264336463313538353131646163366635323362653333326461322f64656d6f732f7376632d73616d706c652e6970796e62&logged_in=true&nwo=Archly2022%2Flogllm&path=demos%2Fsvc-sample.ipynb&platform=mac&repository_id=847660124&repository_type=Repository&version=17#7ae8caf4-3459-4bff-b551-47b20e02b6af