Reasoning integration tests

[mikeperry-tor]

No description provided.

[Copilot]

Pull request overview

Adds integration coverage for “reasoning” responses (including GLM-specific repair scenarios) across multiple proxy/provider integration suites, ensuring both plaintext and E2EE pathways handle reasoning fields and multimodal content decoding correctly.

Changes:

• Introduces shared reasoning integration helpers (runReasoningResponseTests, runGLMReasoningRepairTests) and assertions for both non-stream and SSE streaming responses.
• Adds reasoning/repair subtests for Tinfoil (cloud + direct) and NEAR (nearcloud + neardirect) integration suites.
• Adjusts some provider-specific integration POST helpers to include reasoning_effort control and provider-specific tool-call request shapes.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 6 comments.

File	Description
internal/proxy/integration_tinfoil_test.go	Adds reasoning/repair subtests for Tinfoil and introduces provider-specific chat/tool POST helpers used by the Tinfoil API surface tests.
internal/proxy/integration_reasoning_test.go	New shared reasoning test helpers: posting reasoning requests, GLM repair scenarios, and response/SSE validation.
internal/proxy/integration_neardirect_test.go	Adds GLM reasoning integration test and switches NEAR Direct chat requests to a helper that explicitly sets reasoning_effort.
internal/proxy/integration_nearcloud_test.go	Adds GLM reasoning integration test coverage for NEAR Cloud (plaintext + E2EE, plus repairs).

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.

[Copilot]

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated no new comments.