Add back history and reset subcommand in magics

[akaihola]

• documentation for %config AiMagics.max_history
• manual testing for %config AiMagics.max_history
• manual testing for limiting the maximum number of exchanges to include in the context
• unit tests for all added features
• design decision about context for custom chains

#551 removed the history associated uniquely with the openai-chat provider in magic commands. It also removed the "reset" command to delete said history. Docs were updated to remove mention of the history and the reset command.

This PR adds back the history in magic commands. It also adds the %ai reset subcommand to delete said history. A mention of the history and the reset command are added to docs.

The history transcript maintains the distinction between human and AI messages by wrapping the prompts and responses in HumanMessage and AIMessage objects.

The maximum number of Human/AI message exchanges to include before the new prompt can be configured using %config AiMagics.max_history = <n>, defaulting to 2. The whole history is still kept in memory, so raising this option value will start including older messages in subsequent interactions.

For non-chat providers, human messages are wrapped in pseudo XML <HUMAN>...</HUMAN> tags prepended with AI: or Human: unless there is nothing but the first prompt. This is yet untested.

[dlqqq]

Thank you for opening this PR and for re-implementing message history in the magics! I've left suggestions about 1) bounding the chat history to 2 exchanges at most, and 2) avoiding the pseudo-XML syntax being used for non-chat providers. This is a good stopgap solution for users who want to use history in AI magics as soon as possible.

There are better ways to pass message history in LangChain however. In the future, we will definitely want to rework this logic to use the new LCEL syntax and use the RunnableWithMessageHistory class from langchain_core.runnables.history; see #392.

[dlqqq]

Thank you for your contribution! It's been a week since we've last heard from you, so I'm closing this PR due to inactivity. Please feel free to re-open this PR after addressing the review feedback.

[akaihola]

Thank you for your contribution! It's been a week since we've last heard from you, so I'm closing this PR due to inactivity. Please feel free to re-open this PR after addressing the review feedback.

I don't seem to have permission to re-open. Could you do that?

[krassowski]

Yeah, that's the problem with closing PRs on GitHub.

[akaihola]

Update: I solved the below question myself, see next message.

---

May I ask for advice @dlqqq and @krassowski?

What might be the reason %config doesn't work for max_history?

[akaihola]

I asked:

What might be the reason %config doesn't work for max_history?

Oh, I seem to have forgotten to add

    config=True,

to the traitlet!

[akaihola]

Based on my manual testing, max_history works as expected when using openrouter:openai/gpt-4o as the model. If I prompt the model for my very first question, verbatim, it gives me the question max_history exchanges back.

Interestingly, if I use openrouter:anthropic/claude-3.5-sonnet:beta instead, the model remembers everything! So is there a history layer on OpenRouter side?!?

[akaihola]

I fixed the behavior when max_history == 0, and added unit tests.

[akaihola]

@krassowski & @dlqqq would it make sense to include message history in the context also when invoking custom chains? I haven't touched that part, so it still only sends the prompt entered in the cell:

        if args.model_id in self.custom_model_registry and isinstance(
            self.custom_model_registry[args.model_id], LLMChain
        ):
            # Get the output, either as raw text or as the contents of the 'text' key of a dict
            invoke_output = self.custom_model_registry[args.model_id].invoke(prompt)

[dlqqq]

@akaihola Awesome work! Thank you for including unit test coverage as well. I've left some additional minor feedback below. Once that's addressed, this is ready to be merged. 👍

[akaihola]

@dlqqq I believe I've now addressed your fine review feedback!

[dlqqq]

@akaihola Awesome work, thank you! 🎉