Continuation of #2936:

• Rich-render Python objects, also support Jupyter representations (Dispatch Python output rendering to kino #2954)

• Add autocompletion and other editor features, currently we only offer syntax highlighting (Implement Python intellisense #3133)

• Refactor and unify functionality in relation to multiple language support - Livebook was originally designed with the assumption that external languages would be added via the distribution protocol. However, Pythonx showed us a different approach to integrate other runtimes, which we should now streamline. (Runtime refactorings #3136)

• Change Learn section with one guide for Erlang and another for Python - the Python notebook can include examples of several libraries which we integrate and render by default (Python in Add notebook walking through Python integrations #3150, Erlang tracked in Erlang integration #2060)

• Add support for distribution and FLAME - the goal is to be able to transfer Pythonx objects transparently across nodes. By integrating with FLAME, we can have better control over the objects lifecycle too (Implement remote evaluation and distributed garbage collection pythonx#46, Support pythonx kino_flame#7)

Under consideration:

• Importing and exporting of .ipynb files
• Export .py files (if a notebook only has Python cells)
• Explore integration with Triton and Nx
• Support Apache Arrow via ADBC and integrate with Pandas and Polars (Add query API that returns a python table adbc#146, Add API to read Arrow data from Python back to Elixir adbc#147)