feat: CLI subcommands and flags implemented

1. gpt-cli chat
   • This will launch gpt-cli in chat mode. Your GPT assistant will keep track of your conversation.

Options:
  -m, --max-tokens <MAX_TOKENS>    The maximum number of tokens the response should have. [default: 300]
  -g, --gpt-model <GPT_MODEL>      The GPT model to use. [default: gpt-3.5-turbo]
  -t, --temperature <TEMPERATURE>  The temperature for sampling. [default: 0.2]
  -l, --logging                    Enable logging.
  -h, --help                       Print help

• commands within chat:
  • exit: will terminate the conversation
  • undo: will remove the last request and answer from the chat history..
  • clear: clears the whole chat history. The conversation will start from scratch.

2. gpt-cli one-shot
   • This will allow to perform a one prompt interaction. This mode expects a set --prompt flag.

    Options:
  -p, --prompt <PROMPT>            The prompt for the model.
  -m, --max-tokens <MAX_TOKENS>    The maximum number of tokens the response should have. [default: 300]
  -g, --gpt-model <GPT_MODEL>      The GPT model to use. [default: gpt-3.5-turbo]
  -t, --temperature <TEMPERATURE>  The temperature for sampling. [default: 0.2]
  -l, --logging                    Enable logging.
  -h, --help                       Print help