docs: Multilingual output how-to (prompt-based)

[apexpdl]

Adds a concise how-to guide for Issue #165 explaining how to enforce multilingual output in Concordia agents via prompt-based directives.

Demonstrates:

• Per-agent language setup using goal parameter
• Minimal Python example for prefab_lib
• Notes on prompt design and best practices

[google-cla]

Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

View this failed invocation of the CLA check for more information.

For the most up to date status, view the checks section at the bottom of the pull request.

[apexpdl]

I signed it

[locross93]

Hi @apexpdl thank you for working on this issue and for writing up this doc!

In this example, do you also want the intermediate prompts within the agent (ie. "what kind of person is {agent_name}?" to be in the target language? If so, modifying the goal string only may not be enough as only the components that have the goal string in their prefix will have the language directive. Or is the intended use case here just the actions or dialog the agent outputs? Can you modify the doc to make the distinction here clear. Other than that this looks good.

[apexpdl]

Hey @locross93 ,Thanks for the catch . I updated the doc to clarify that the language directive in this example is intended for user-facing outputs (dialog/actions), and added a note that intermediate/internal prompts may not inherit the directive when only goal is modified, plus guidance for enforcing language more broadly if needed.