Refactor hackett

[augeorge]

• started refactoring hackett with the new pytensor from data function
• wrote function to convert enzyme data file (for hackett) to dataclass expected by new function
• wrote tests for the above conversion function
• compared the shape of the original and refactored pytensors (log_en_t) are the same
• calculated prior predictives for original and refactored (100 samples)
• compared the resulting prior predictive for chi_obs is the same for the median, mean, and stdev distributions

[augeorge]

still need to:

• compare the prior predictives for vn_obs and log_en_t
• check the ordering of metabolites for chi_obs
• refactor hackett to use the noisy obs function
• rerun comparisons
• make a comparison integration test