fix(openai): omit prompt_cache_key for openai-compatible providers

[gnanam1990]

Summary

• Interactive Zero always has a session ID, so completion requests were sending OpenAI’s prompt_cache_key.
• Strict openai-compatible gateways (e.g. NVIDIA NIM) reject that field with Unsupported parameter(s): prompt_cache_key, while plain zero exec usually has no session and works.
• This PR disables prompt_cache_key for ProviderKindOpenAICompatible and keeps it for official OpenAI cache routing.
• ZERO_DISABLE_PROMPT_CACHE_KEY=1 remains a global kill switch.

Fixes #624

Changes

• internal/providers/openai/provider.go — add DisablePromptCacheKey option and gate request mapping
• internal/providers/factory.go — set the flag for openai-compatible profiles only
• Tests covering OpenAI vs openai-compatible serialization

Test plan

• go test ./internal/providers/openai/ ./internal/providers/ -count=1 -run 'TestOpenAIRequestPromptCacheKey|TestPromptCacheKeyOnlyOnOfficialOpenAI'
• Interactive zero against an openai-compatible provider (e.g. NIM) — send a prompt, no prompt_cache_key validation error
• Official OpenAI multi-turn still receives prompt_cache_key when a session is present
• ZERO_DISABLE_PROMPT_CACHE_KEY=1 still forces omit on OpenAI

Summary by CodeRabbit

• Bug Fixes

  • OpenAI-compatible gateways no longer receive the OpenAI-specific prompt_cache_key field, preventing request rejections.
  • Official OpenAI requests continue to include the prompt cache key when provided.
  • Existing environment-based controls for disabling prompt cache keys remain supported.

• Tests

  • Added coverage verifying prompt cache key behavior for official OpenAI and compatible gateways.

[github-actions]

Zero automated PR review

Verdict: No blockers found

Blockers

• None found.

Validation

• [pass] Diff hygiene: git diff --check
• [pass] Tests: go test ./...
• [pass] Build: go run ./cmd/zero-release build
• [pass] Smoke build: go run ./cmd/zero-release smoke

Scope

Head: 346776be01e7
Changed files (5): internal/providers/factory.go, internal/providers/factory_test.go, internal/providers/openai/provider.go, internal/providers/openai/provider_test.go, internal/providers/openai/types.go

This deterministic review checks validation status and basic diff hygiene. A human reviewer still owns product judgment and design quality.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

Run ID: 71b2a8de-f044-4ede-ab94-86e877850cd6

📥 Commits

Reviewing files that changed from the base of the PR and between dc06fe7 and 346776b.

📒 Files selected for processing (5)

• internal/providers/factory.go
• internal/providers/factory_test.go
• internal/providers/openai/provider.go
• internal/providers/openai/provider_test.go
• internal/providers/openai/types.go

---

Walkthrough

Changes

Prompt cache key routing

Layer / File(s)	Summary
Provider option wiring internal/providers/openai/provider.go, internal/providers/openai/types.go, internal/providers/factory.go	Adds DisablePromptCacheKey and enables it when constructing OpenAI-compatible providers.
Request filtering and validation internal/providers/openai/provider.go, internal/providers/openai/provider_test.go, internal/providers/factory_test.go	Omits prompt_cache_key for compatible providers while retaining it for official OpenAI, with provider and factory tests covering serialization.

Estimated code review effort: 3 (Moderate) | ~20 minutes

Possibly related PRs

• Gitlawb/zero#515: Related OpenAI provider prompt-cache key forwarding and kill-switch behavior.

Suggested reviewers: Vasanthdev2004, kevincodex1

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 66.67% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the main change: omitting prompt_cache_key for OpenAI-compatible providers.
Linked Issues check	✅ Passed	The changes address #624 by disabling prompt_cache_key for OpenAI-compatible providers and adding coverage for the behavior.
Out of Scope Changes check	✅ Passed	The remaining changes are supporting code and tests for the same prompt_cache_key behavior, with no clear unrelated additions.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/omit-prompt-cache-key-compatible

---

_{Comment @coderabbitai help to get the list of available commands.}