fix(workflow): preserve non-ASCII characters in LLM agent node input

[anneheartrecord]

Summary

input_schema inputs containing non-Latin characters (Hebrew, Chinese, etc.) reach the LLM as \uXXXX escapes, which bloats prompt tokens (~6x for Hebrew) and degrades model responses, as reported in #6279.

The escaping comes from json.dumps being called with its default ensure_ascii=True on the LLM-bound input text. Two paths were affected:

• workflow/_llm_agent_wrapper.py – _node_input_to_content() for dict/list node input.
• flows/llm_flows/contents.py – _build_task_input_user_content(), which rebuilds a delegated task's function-call args as the sub-agent's first user turn. This path takes priority over the wrapper fallback, so it affects the common chat/root → task sub-agent delegation case.

Both now serialize with ensure_ascii=False, matching how the output-schema path already serializes responses (_output_schema_processor.py), fixed earlier in #2936/#2937.

Note: the BaseModel branch (model_dump_json()) is intentionally left unchanged — Pydantic v2 already emits raw UTF-8 (does not escape non-ASCII), which the added regression test confirms.

Testing

• tests/unittests/workflow/test_llm_agent_as_node.py – dict/list/BaseModel node input preserves non-ASCII.
• tests/unittests/flows/llm_flows/test_contents.py – delegated task FC args preserve non-ASCII.

Both new tests fail before the change and pass after.

Fixes #6279

[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.