fix(byok): support user OpenAI/Anthropic API keys + working effort pill

[OmGuptaIND]

Summary

Three connected bugs that surfaced when a user supplies their own OpenAI/Anthropic key (BYOK) instead of using the Anton/OpenRouter proxy:

1. resolveModel rejected BYOK models. Only anton and openrouter had fallbacks — direct-API providers (openai, anthropic, google, groq, mistral) relied entirely on pi SDK's hardcoded registry, so any model the registry didn't know (e.g. gpt-5.5) failed to start a session. New buildDirectApiModel mirrors buildOpenRouterModel for these providers and routes to the right api/baseUrl. resolveModel now also threads an optional baseUrl override so users with self-hosted/proxy endpoints work.

2. Mid-conversation switchModel was brittle. It called pi SDK's setModel but didn't reconcile the rest of the model-derived state. Hardened to (a) fail fast with a clear error when the new provider has no API key, (b) clamp thinking level to 'off' on reasoning → non-reasoning transitions so the upstream API doesn't reject the request, (c) refresh compactionConfig.maxContextTokens so the gauge and compaction threshold match the new model's window.

3. Effort pill was hidden for API sessions on gpt-5.x. Pi SDK and the server-side wiring were already correct — Session.setThinkingLevel forwards to piAgent.setThinkingLevel. The composer's supportsReasoningEffort() regex just didn't include the gpt-5 family. Worked for the codex harness because that branch returns true unconditionally. Regex now matches the same families buildDirectApiModel detects, with a comment pointing the two heuristics at each other so they don't drift.

Test plan

• Add an OpenAI API key in Provider settings, pick gpt-5.5, send a message → session starts, response streams.
• Add an Anthropic API key, pick claude-sonnet-4-6, send a message → works.
• Mid-conversation, switch from claude-opus-4-6 (reasoning) → a non-reasoning model → next turn doesn't get rejected for sending a thinkingLevel.
• Mid-conversation, switch to a provider with no key configured → clear error in the log; other sessions keep working.
• With provider=openai, model=gpt-5.5 selected, the Effort pill is visible in the composer and changes the next turn's effort.
• With provider=codex (harness), the Effort pill still shows.
• With provider=claude (Claude Code harness), the Effort pill stays hidden (CLI has no thinking flag).

🤖 Generated with Claude Code