fix(openai): add support for logprobs when using the response api

[HOZHENWAI]

Fixes #36211

Fix the issue where the combination of use_responses_api=True and logprobs=True prevented any invocation.
Precommit validated and all unit-tests passing including the new ones written for this fix.
Test scope are discutable: i've decided to test that the request_payload is compatible with the response api instead of testing the output of _construct_responses_api_payload in the case of logprobs=True which may have been more proper. If necessary i can do the changes. Also, i have no tested the case of refusal when logprobs = True.
I also think a small update to the integration test could be useful but did not write it.
Final test made was to make sure that for the following snipped
`from langchain_openai import ChatOpenAI

client = ChatOpenAI(
model="gpt-5.2",
use_responses_api=True,
logprobs=True,
)

response = client.invoke("say hello world")`
response had the logprobs in the response_metadata which is the case with this PR.

Read the full contributing guidelines: https://docs.langchain.com/oss/python/contributing/overview

All contributions must be in English. See the language policy.

If you paste a large clearly AI generated description here your PR may be IGNORED or CLOSED!

Thank you for contributing to LangChain! Follow these steps to have your pull request considered as ready for review.

1. PR title: Should follow the format: TYPE(SCOPE): DESCRIPTION

• Examples:
  • fix(anthropic): resolve flag parsing error
  • feat(core): add multi-tenant support
  • test(openai): update API usage tests
• Allowed TYPE and SCOPE values: https://github.com/langchain-ai/langchain/blob/master/.github/workflows/pr_lint.yml#L15-L33

2. PR description:

• Write 1-2 sentences summarizing the change.
• The Fixes #xx line at the top is required for external contributions — update the issue number and keep the keyword. This links your PR to the approved issue and auto-closes it on merge.
• If there are any breaking changes, please clearly describe them.
• If this PR depends on another PR being merged first, please include "Depends on #PR_NUMBER" in the description.

3. Run make format, make lint and make test from the root of the package(s) you've modified.

• We will not consider a PR unless these three are passing in CI.

4. How did you verify your code works?

Additional guidelines:

• All external PRs must link to an issue or discussion where a solution has been approved by a maintainer, and you must be assigned to that issue. PRs without prior approval will be closed.
• PRs should not touch more than one package unless absolutely necessary.
• Do not update the uv.lock files or add dependencies to pyproject.toml files (even optional ones) unless you have explicit permission to do so by a maintainer.

Social handles (optional)

Twitter: @
LinkedIn: https://linkedin.com/in/

[codspeed-hq]

Merging this PR will improve performance by 13%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚡ 3 improved benchmarks
⏩ 37 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	WallTime	test_init_time	3.6 ms	3.2 ms	+11.88%
⚡	WallTime	test_init_time	3.6 ms	3.2 ms	+13%
⚡	WallTime	test_init_time	136 ms	122.1 ms	+11.36%

---

_{Comparing HOZHENWAI:HOZHENWAI/fix-missing-logprobs-openai-response (9921b2c) with master (342d8bd)}

Footnotes

1. 37 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[ccurme (ccurme)]

should this go in response_metadata, or on the associated content block?

[HOZHENWAI]

well, is that a recommandation or a question? The default pattern without the response api when using logprobs is to have it in the response_metadata. I just copied the pattern to make sure the outputs were compatible between use_responses_api=True and use_responses_api=False . Should this be changed?
Tbh, I don't think I tested the case with multi-turn conversation when logprobs=True and use_response_api=False, so I don't know if the pattern is different in that case.