Resolves #1

Why the bug occurs

The ParallelEpisode class is unhashable because it does not implement the __hash__ and __eq__ methods. This causes a TypeError when attempting to add instances of ParallelEpisode to a set or use them as dictionary keys in the recognize_candidate_parallel function.

How to reproduce

1. Use the WINEPI library on your own dataset.
2. Call the methods w.discover_frequent_episodes() or w.generate_rules().
3. Observe that empty arrays are returned.
4. Alternatively, run the example code from the README.md, which triggers the following error:

TypeError: unhashable type: 'ParallelEpisode'

How to fix

Implement the __hash__ and __eq__ methods in the ParallelEpisode class to make its instances hashable. This allows instances to be added to sets and used as dictionary keys without raising a TypeError. Here's a step-by-step approach:

1. Define __eq__ Method: Ensure that two ParallelEpisode instances are considered equal if their relevant attributes are equal.
2. Define __hash__ Method: Compute a hash value based on the same attributes used in the __eq__ method.
3. Update recognize_candidate_parallel Function: Ensure that the function correctly utilizes the hashable ParallelEpisode instances.

This fix will allow the WINEPI functions to work as expected without encountering hash-related errors.

Test these changes locally

git checkout -b gitauto/issue-1-20241220-000416
git pull origin gitauto/issue-1-20241220-000416