Add CNN class for handling MNIST dataset

[sweep-nightly]

Description

This PR adds a new CNN class in the cnn.py file to handle the MNIST dataset. The CNN class is implemented using PyTorch and includes convolutional layers, pooling layers, and fully connected layers. Additionally, a train_cnn function is defined to train the CNN on the MNIST data.

Summary of Changes

• Created a new file src/cnn.py to contain the CNN class and the train_cnn function.
• Imported necessary PyTorch modules in cnn.py.
• Defined a new class CNN in cnn.py that inherits from torch.nn.Module.
• Implemented the __init__ method in the CNN class to define the layers of the CNN.
• Implemented the forward method in the CNN class to perform the forward pass of the CNN.
• Defined the train_cnn function in cnn.py to train the CNN on the MNIST data.
• Added comments and docstrings to explain the functionality of the CNN class and the train_cnn function.
• Imported the CNN class and the train_cnn function in src/main.py.
• Created an instance of the CNN class in main.py after loading and preprocessing the MNIST data.
• Called the train_cnn function with the CNN instance, DataLoader instance, and number of epochs in main.py.

Please review and merge this PR to incorporate the new CNN class for handling the MNIST dataset.

Fixes #9.

---

🎉 Latest improvements to Sweep:

• Sweep can now passively improve your repository! Check out Rules to learn more.

---

💡 To get Sweep to edit this pull request, you can:

• Comment below, and Sweep can edit the entire PR
• Comment on a file, Sweep will only modify the commented file
• Edit the original issue to get Sweep to recreate the PR from scratch

[sweep-nightly]

Rollback Files For Sweep

• Rollback changes to src/cnn.py
• Rollback changes to src/main.py

[sweep-nightly]

Apply Sweep Rules to your PR?

• Apply: All docstrings and comments should be up to date.
• Apply: Code should be properly formatted and indented.
• Apply: Variable and function names should be descriptive and follow a consistent naming convention.
• Apply: Imports should be organized and grouped together.
• Apply: There should be no unused imports or variables.
• Apply: Code should be properly commented and include docstrings for functions and classes.