Python: Fix structured value parsing for split text chunks

[t-anjan]

Structured response parsing currently uses the public text property, which inserts spaces between TextContent chunks inside a Message. That can corrupt split JSON structured outputs while leaving normal display text behavior correct.

This change keeps Message.text unchanged and uses a structured-output-only text path that concatenates text chunks without synthetic spaces before parsing ChatResponse.value and AgentResponse.value.

Tests:

• cd python && uv run pytest packages/core/tests/core/test_types.py
• cd python && uv run ruff check packages/core/agent_framework/_types.py packages/core/tests/core/test_types.py
• git diff --check

[Copilot]

Pull request overview

Fixes structured-output parsing for ChatResponse.value and AgentResponse.value by avoiding the synthetic spaces introduced by Message.text when a message contains multiple text Content chunks (which can corrupt split JSON). The public Message.text/Response.text behavior remains unchanged for display.

Changes:

• Added structured-output-only text assembly helpers that concatenate text chunks without inserting spaces.
• Updated ChatResponse.value and AgentResponse.value to parse structured output from the structured-only text path instead of self.text.
• Added regression tests covering split JSON across text chunks while preserving Message.text spacing behavior.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
python/packages/core/agent_framework/_types.py	Adds structured-only text helpers and switches structured parsing to use them (avoids Message.text synthetic spacing).
python/packages/core/tests/core/test_types.py	Adds tests ensuring split structured JSON parses correctly without changing Message.text/response.text.

[github-actions]

Python Test Coverage Report •

File	Stmts	Miss	Cover	Missing
packages/core/agent_framework
_types.py	1226	97	92%	61, 70–71, 125, 130, 149, 151, 155, 159, 161, 163, 165, 183, 187, 213, 235, 240, 245, 249, 279, 705–706, 887–888, 1346, 1421, 1456, 1476, 1486, 1538, 1672–1674, 1959–1964, 1989, 2044, 2049, 2059, 2067, 2074–2078, 2096, 2169, 2182, 2187, 2300, 2323, 2584, 2608, 2707, 2888–2889, 2991, 3220, 3273, 3292, 3331, 3342, 3344–3348, 3350, 3353–3361, 3371, 3460, 3597, 3602, 3607, 3612, 3616, 3702–3704, 3733, 3821–3825
TOTAL	44418	5243	88%

Python Unit Test Overview

Tests	Skipped	Failures	Errors	Time
8959	33 💤	0 ❌	0 🔥	2m 32s ⏱️

[moonbox3]

Could you please resolve the conflicts?

[t-anjan]

Resolved the conflicts with current main in _types.py and test_types.py.

Upstream changed structured-output parsing to use the final non-empty assistant message, while this PR fixes parsing of structured JSON split across multiple text content chunks (without the spaces inserted by Message.text). The resolution keeps upstream’s final-message selection and makes that helper concatenate the selected message’s raw text chunks directly. I retained regression coverage for both behaviors; uv run --directory packages/core pytest tests/core/test_types.py -q passes all 278 tests.