Implement missing endpoints in Inference service

[lcalvobartolome]

The missing endpoints are:

• InferCorpus
• listInferenceModels
• deleteInferenceModel

[lcalvobartolome]

Important Considerations:

• Ignore the SparkLDA inferencer for now.
• When implementing the inferencer for CTM in Python, you may need to modify lines 471-472 in the inferencer/src/core/inferencer/base/inferencer.py file. Rrefer to lines 1446-1449 in the topicmodeling.py file from the topicmodeler repository for further guidance.
• When inferring topics for a single document provided by the user using CTM, embeddings are not required. However, if you are inferring topics for a whole corpus, the embeddings should be provided in the input parquet file of the corpus for inference.

TODO:

1. Construct ProdLDA model for a different corpus (e.g., SCOPUS).
2. Index both corpus and model into the EWB.
3. Implement inferDoc functionality for ProdLDA and test it with the former model.
4. Repeat 1-3 for CTM.
5. Implement listInferenceModels and deleteInferenceModel.
6. Implement InferCorpus.