Add ClickHouse integration with writer service and configuration files

delete: Remove unused model files from Dockerfile.consumer

Add incident suggestion functionality to Qwen model integration

delete: Remove unused Dockerfile for AWS log poller Loki

Delete llm(fine_tuning).py

Added the LLM fine-tuning file

Delete LLM(Fine_tuning).ipynb

Added the Google Colab LLM fine-tuning file

fix(docs): Correct formatting of the pipeline explanation section in README

feat(docs): Update README to reflect in-house model usage and enhance fine-tuning and Loki integration sections

Merge branch LLMtest

feat(ml): Replace instruction-tuned model with our own fine-tuned log classifier

loki1

loki

feat: Enhance fine-tuning script with checkpoint resume, model merging, and Hugging Face upload