fix(opencode): keep Claude models off OpenAI provider

## Description

The injected `headroom` OpenCode provider uses
`@ai-sdk/openai-compatible` and the proxy's `/v1/chat/completions`
route. It currently advertises Claude model IDs in that provider, so
OpenCode sends Claude requests to the OpenAI upstream and receives
`invalid_api_key` errors. Keep Claude on OpenCode's native `anthropic`
provider, which Headroom already redirects to the proxy.

Closes #2911

## Type of Change

- [x] Bug fix (non-breaking change that fixes an issue)
- [ ] New feature (non-breaking change that adds functionality)
- [ ] Breaking change (bug fix or feature that would cause existing
functionality to change)
- [ ] Documentation update
- [ ] Performance improvement
- [ ] Code refactoring (no functional changes)

## Changes Made

- Remove Claude IDs from the injected OpenAI-compatible provider model
map.
- Keep GPT models available through the `headroom/<id>` namespace.
- Add regression assertions that generated config never advertises
Claude models on this endpoint.

## Testing

- [x] Unit tests pass (`pytest`)
- [x] Linting passes (`ruff check .`)
- [ ] Type checking passes (`mypy headroom`)
- [x] New tests added for new functionality
- [x] Manual testing performed

### Test Output

```text
python -m pytest tests/test_providers_opencode_config.py -q -k "not build_launch_env_with_project"
40 passed, 1 deselected

python -m ruff check headroom/providers/opencode/config.py tests/test_providers_opencode_config.py
All checks passed!

python -m compileall -q headroom/providers/opencode/config.py tests/test_providers_opencode_config.py
(pass)
```

The full config test module also exposes an unrelated pre-existing
Windows path assertion failure in `test_build_launch_env_with_project`;
the failure is caused by comparing a native `Path` string with
JSON-escaped backslashes and is outside this change.

## Real Behavior Proof

- Environment: Windows 11, Python 3.11; no external API credentials
used.
- Exact command / steps: `python -c "from
headroom.providers.opencode.config import headroom_provider_entry;
print(sorted(headroom_provider_entry(8787)['models']))"`
- Observed result: `['gpt-4.1', 'gpt-4o']`; the generated
OpenAI-compatible provider no longer advertises any `claude-*` IDs.
- Not tested: live OpenCode request routing or a vendor API call,
because they require external credentials. The regression suite verifies
the generated configuration consumed by OpenCode.

## Review Readiness

- [x] I have performed a self-review
- [x] This PR is ready for human review

## Checklist

- [x] My code follows the project's style guidelines
- [x] I have performed a self-review of my code
- [ ] I have commented my code, particularly in hard-to-understand areas
(not needed; the provider routing rationale is documented inline)
- [ ] I have made corresponding changes to the documentation (the
generated provider behavior is documented in code; existing docs
describe the separate npm provider)
- [x] My changes generate no new warnings
- [x] I have added tests that prove my fix is effective
- [x] New and existing relevant unit tests pass locally with my changes
- [x] I did **not** edit `CHANGELOG.md`

## Additional Notes

The native `anthropic` and `openai` provider entries both continue to
point at the Headroom proxy, so this change only removes an invalid
duplicate Claude route and does not affect native Claude traffic.

headroom/providers/opencode/config.py

@@ -31,24 +31,18 @@ _MCP_BLOCK_RE = re.compile(
 )
 HEADROOM_OPENCODE_PLUGIN = "headroom-opencode"
 
-# Models exposed by the injected `headroom` provider. OpenCode only resolves
-# `headroom/<id>` for ids listed in the provider's `models` map, so an empty
-# map means every documented `headroom/*` model fails with "Model not found".
-# Keep in sync with DEFAULT_MODELS in plugins/opencode/src/provider.ts and the
-# table in plugins/opencode/README.md.
+# Models exposed by the injected `headroom` provider. This provider uses
+# ``@ai-sdk/openai-compatible`` and the proxy's ``/v1/chat/completions`` path,
+# which is routed to the configured OpenAI upstream. Do not advertise Claude
+# models here: OpenCode would send them through the OpenAI endpoint and report
+# an ``invalid_api_key`` error instead of reaching Anthropic. Claude models are
+# available through OpenCode's native ``anthropic`` provider, whose base URL is
+# also redirected to Headroom by ``build_opencode_config_content``.
+#
+# OpenCode only resolves ``headroom/<id>`` for ids listed in this map, so an
+# empty map means every documented ``headroom/*`` model fails with "Model not
+# found".
 HEADROOM_OPENCODE_MODELS: dict[str, Any] = {
-    "claude-sonnet-4-6": {
-        "name": "Claude Sonnet 4.6",
-        "limit": {"context": 200000, "output": 16384},
-    },
-    "claude-opus-4-6": {
-        "name": "Claude Opus 4.6",
-        "limit": {"context": 200000, "output": 16384},
-    },
-    "claude-haiku-4-5-20251001": {
-        "name": "Claude Haiku 4.5",
-        "limit": {"context": 200000, "output": 8192},
-    },
     "gpt-4o": {
         "name": "GPT-4o",
         "limit": {"context": 128000, "output": 16384},

tests/test_providers_opencode_config.py

@@ -199,7 +199,10 @@ def test_inject_provider_config_creates_file(
     assert config["provider"]["headroom"]["npm"] == "@ai-sdk/openai-compatible"
     # Bare model ids: OpenCode resolves them as "headroom/<id>" (#1657).
     models = config["provider"]["headroom"]["models"]
-    assert "claude-sonnet-4-6" in models
+    assert set(models) == {"gpt-4o", "gpt-4.1"}
+    # The injected provider is OpenAI-compatible. Claude models must remain on
+    # OpenCode's native Anthropic provider so they are not sent to OpenAI.
+    assert not any(model_id.startswith("claude-") for model_id in models)
     assert all(not model_id.startswith("headroom/") for model_id in models)
     assert "mcp" not in config
     assert "model" not in config  # headroom provider is a transparent pass-through
@@ -456,10 +459,12 @@ def test_build_opencode_config_content_without_mcp(
     providers = config["provider"]
     assert providers["anthropic"]["options"]["baseURL"] == "http://127.0.0.1:8787/v1"
     assert providers["openai"]["options"]["baseURL"] == "http://127.0.0.1:8787/v1"
-    # The headroom provider exposes explicit models so "headroom/<id>" resolves (#1657).
+    # The headroom provider exposes only models supported by its
+    # OpenAI-compatible endpoint so "headroom/<id>" resolves safely (#1657).
     assert providers["headroom"]["options"]["baseURL"] == "http://127.0.0.1:8787/v1"
     models = providers["headroom"]["models"]
-    assert "claude-sonnet-4-6" in models
+    assert set(models) == {"gpt-4o", "gpt-4.1"}
+    assert not any(model_id.startswith("claude-") for model_id in models)
     assert all(not model_id.startswith("headroom/") for model_id in models)
     # The transport plugin is injected by absolute path (opencode loads it directly).
     assert config["plugin"] == [str(plugin)]