feat: Create Jupyter notebook for OSRS RL training

[ConjureGanja]

This commit introduces a new Jupyter notebook, rl_training_osrs.ipynb, designed to guide you through setting up and running reinforcement learning training for an OSRS agent using this repository.

The notebook includes the following sections:

• Project Structure: Overview of the key directories and their contents.
• Setting up the Conda Environment: Instructions for creating and activating the necessary Conda environment, which includes PyTorch and Java 17.
• Launching the OSRS Simulation Server: Steps to run the Elvarg RSPS based simulation environment.
• Starting a Training Job: How to initiate an RL training process using predefined configurations.
• Monitoring Training with TensorBoard: Instructions for visualizing training metrics.
• Evaluating a Trained Model: How to run a trained agent in the simulation.
• Serving Models via API: Information on deploying models as an API service.
• Conclusion and Next Steps: Summary and suggestions for further exploration.

The notebook aims to provide a clear and comprehensive guide for you to get started with the OSRS RL project. Content was derived from existing README files and organized into a step-by-step tutorial format.

[Copilot]

Pull Request Overview

This PR adds a new Jupyter notebook (rl_training_osrs.ipynb) that walks users through setting up, training, evaluating, and serving an OSRS RL agent using the repository.

• Introduces step-by-step instructions in a notebook format
• Covers environment setup, simulation server launch, and training workflow
• Includes sections on monitoring, evaluation, and serving models via API

Comments suppressed due to low confidence (3)

rl_training_osrs.ipynb:7

• [nitpick] Consider adding a 'Prerequisites' section before the project structure to specify required Python and Java versions, OS support, and any other system dependencies.

## 1. Project Structure

rl_training_osrs.ipynb:166

• Clarify the default location of TensorBoard log files and ensure consistency with the earlier mention of logs stored in ./logs and ./experiments/<name>.

This command specifically looks for TensorBoard logs generated by the training framework.

rl_training_osrs.ipynb:89

• Add a note that on Windows systems the user should run gradlew.bat run instead of ./gradlew run to avoid confusion.

./gradlew run

[ConjureGanja]

Seems so far so good. Has the added steps for setting up environment and launching scripts.