fix(ai): skip errored/aborted assistant messages in transform-messages

Fixes OpenAI Responses 400 error 'reasoning without following item' by
skipping errored/aborted assistant messages entirely rather than filtering
at the provider level. This covers openai-responses, openai-codex-responses,
and future providers.

Removes strictResponsesPairing compat option (no longer needed).

Closes #838

packages/coding-agent/CHANGELOG.md

@@ -14,7 +14,7 @@
 
 ### Added
 
-- Added `strictResponsesPairing` compat option for custom OpenAI Responses models on Azure ([#768](https://github.com/badlogic/pi-mono/pull/768) by [@nicobako](https://github.com/nicobako))
+- Added `strictResponsesPairing` compat option for custom OpenAI Responses models on Azure ([#768](https://github.com/badlogic/pi-mono/pull/768) by [@prateekmedia](https://github.com/prateekmedia))
 - Session selector (`/resume`) now supports path display toggle (`Ctrl+P`) and session deletion (`Ctrl+D`) with inline confirmation ([#816](https://github.com/badlogic/pi-mono/pull/816) by [@w-winter](https://github.com/w-winter))
 - Added undo support in interactive mode with Ctrl+- hotkey. ([#831](https://github.com/badlogic/pi-mono/pull/831) by [@Perlence](https://github.com/Perlence))
 

packages/coding-agent/README.md

@@ -751,14 +751,6 @@ To fully replace a built-in provider with custom models, include the `models` ar
 | `supportsUsageInStreaming` | Whether provider supports `stream_options: { include_usage: true }`. Default: `true` |
 | `maxTokensField` | Use `max_completion_tokens` or `max_tokens` |
 
-**OpenAI Responses (`openai-responses`):**
-
-| Field | Description |
-|-------|-------------|
-| `strictResponsesPairing` | Enforce strict reasoning/message pairing when replaying OpenAI Responses history on providers like Azure (default: `false`) |
-
-If you see 400 errors like "item of type 'reasoning' was provided without its required following item" or "message/function_call was provided without its required reasoning item", set `compat.strictResponsesPairing: true` on the affected model in `models.json`.
-
 **Live reload:** The file reloads each time you open `/model`. Edit during session; no restart needed.
 
 **Model selection priority:**

packages/coding-agent/src/core/model-registry.ts

@@ -28,7 +28,7 @@ const OpenAICompletionsCompatSchema = Type.Object({
 });
 
 const OpenAIResponsesCompatSchema = Type.Object({
-	strictResponsesPairing: Type.Optional(Type.Boolean()),
+	// Reserved for future use
 });
 
 const OpenAICompatSchema = Type.Union([OpenAICompletionsCompatSchema, OpenAIResponsesCompatSchema]);