feat(ai): add strictResponsesPairing for Azure OpenAI Responses API

Split OpenAICompat into OpenAICompletionsCompat and OpenAIResponsesCompat
for type-safe API-specific compat settings. Added strictResponsesPairing
option to suppress orphaned reasoning/tool calls on incomplete turns,
fixing 400 errors on Azure's Responses API which requires strict pairing.

Closes #768

packages/coding-agent/CHANGELOG.md

@@ -2,6 +2,10 @@
 
 ## [Unreleased]
 
+### 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))
+
 ### Changed
 
 - Share URLs now use hash fragments (`#`) instead of query strings (`?`) to prevent session IDs from being sent to buildwithpi.ai ([#828](https://github.com/badlogic/pi-mono/issues/828))

packages/coding-agent/README.md

@@ -735,6 +735,8 @@ To fully replace a built-in provider with custom models, include the `models` ar
 
 **OpenAI compatibility (`compat` field):**
 
+**OpenAI Completions (`openai-completions`):**
+
 | Field | Description |
 |-------|-------------|
 | `supportsStore` | Whether provider supports `store` field |
@@ -743,6 +745,14 @@ 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

@@ -20,13 +20,19 @@ import type { AuthStorage } from "./auth-storage.js";
 const Ajv = (AjvModule as any).default || AjvModule;
 
 // Schema for OpenAI compatibility settings
-const OpenAICompatSchema = Type.Object({
+const OpenAICompletionsCompatSchema = Type.Object({
 	supportsStore: Type.Optional(Type.Boolean()),
 	supportsDeveloperRole: Type.Optional(Type.Boolean()),
 	supportsReasoningEffort: Type.Optional(Type.Boolean()),
 	maxTokensField: Type.Optional(Type.Union([Type.Literal("max_completion_tokens"), Type.Literal("max_tokens")])),
 });
 
+const OpenAIResponsesCompatSchema = Type.Object({
+	strictResponsesPairing: Type.Optional(Type.Boolean()),
+});
+
+const OpenAICompatSchema = Type.Union([OpenAICompletionsCompatSchema, OpenAIResponsesCompatSchema]);
+
 // Schema for custom model definition
 const ModelDefinitionSchema = Type.Object({
 	id: Type.String({ minLength: 1 }),