feat: Generate command

[carlosgjs]

This PR builds upon #5

• Support for autodoc generate <python_file>
• Renamed predict to eval since it'll be used for computing metrics
• AzureML test run for generate.yml

Sample local run:

$ autodoc generate AOS/sweetPeaEnglishTranslator/validation/codeToCode/static_text/example_1/code_1.py
2023-12-05 16:43:56,498 INFO predict_hf.__init__(): Loading model from meta-llama/llama-2-7b-chat-hf
Loading checkpoint shards: 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 2/2 [00:04<00:00,  2.21s/it]
2023-12-05 16:44:19,785 INFO predict_hf.__init__(): Model loaded
2023-12-05 16:44:20,162 INFO predict_hf.predict(): Generating 1 predictions
2023-12-05 16:46:05,671 INFO predict_hf.predict(): Generated 1 results
2023-12-05 16:46:05,671 INFO main.generate(): Writing output to output.txt

$ cat output.txt
Purpose:
The purpose of this experiment is to investigate the relationship between word recognition and color perception. Specifically, the experiment aims to determine whether the recognition of a word is influenced by the color of the ink in which it is written.

Procedure:
The experiment consists of a within-subjects design, in which participants are presented with a series of words written in different colors. The participants are asked to recognize the words and indicate whether they are congruent or incongruent with the color of the ink. The experiment uses a blocking procedure, in which the order of the words is randomized within each participant, and the color of the ink is constant for each word. The experiment also includes a minimum number of trials constraint to ensure that the results are reliable and accurate.

[younesStrittmatter]

looks good