Add DeepSeek R1 intregration

[sjanaX01]

DeepSeek R1 is amazing. Add DeepSeek R1 to sgpt.

[Abyss-c0re]

After pip install litellm I was able to get it running locally with this config.

CHAT_CACHE_PATH=/tmp/chat_cache
CACHE_PATH=/tmp/cache
CHAT_CACHE_LENGTH=100
CACHE_LENGTH=100
REQUEST_TIMEOUT=60
DEFAULT_MODEL=ollama/deepseek-r1:14b
DEFAULT_COLOR=magenta
ROLE_STORAGE_PATH=~/.config/shell_gpt/roles
DEFAULT_EXECUTE_SHELL_CMD=false
DISABLE_STREAMING=false
CODE_THEME=dracula
OPENAI_FUNCTIONS_PATH=~/.config/shell_gpt/functions
OPENAI_USE_FUNCTIONS=false
SHOW_FUNCTIONS_OUTPUT=false
API_BASE_URL=http://127.0.0.1:11434
PRETTIFY_MARKDOWN=true
USE_LITELLM=true
SHELL_INTERACTION=true
OS_NAME=auto
SHELL_NAME=auto
OPENAI_API_KEY=AA

Since it's spamming errors and and lacking proper support I made a simple python wrapper myself. Maybe it will evolve into something like this project.

https://github.com/catoni0/deepseek_shell