epic: Cortex "App Shell" for v0.1

[dan-menlo]

Goal

• Cortex has an "app shell"
  • e.g. what processes do we spawn?
  • e.g. where is data persisted?
• "App shell" is well documented for internal team + users

Specs

As we work on this epic, we should capture key decisions in the evergreen App Shell architectural spec:

• Cortex.cpp: Data Structures #1115
• Spec: Cortex.cpp Logs #1112

Tasklist

Deps

• chore: Improve package/binary size by remove jinja2 #1063

Processes

• Start process in background #1070
• Feat background process server #1043

Data Structures

• feat: update data folder path #1121
• feat: Read application config .cortexrc from file #1031

Docs:

Key Decisions should be documented here:

• https://github.com/janhq/cortex.cpp/discussions/1150
• Cortex.cpp: Data Structures #1115

Things to think about

• Why are there separate processes for Server and CLI?
  • Should we separate logs for Server and CLI?

[vansangpfiev]

@dan-homebrew Can you elaborate on app shell?. We are starting server in background and use CLI/other to request. Does this mean instead of running server in background, we will start it as a daemon?