feat(providers): add xAI provider with device code OAuth

Add a separate `xai` provider type for xAI Grok CLI subscription accounts.
It is independent of the existing `grok` provider, which reverse-proxies
grok.com with browser cookies; behavior of `grok` is unchanged.

Account binding uses the xAI device code flow, so no local callback
listener is needed and headless deployments can bind accounts. Refresh
tokens can also be imported individually or in batches, and are rotated
on refresh.

OAuth requests default to the cli-chat-proxy Responses API; API keys and
compact stay on api.x.ai. Explicit custom gateways are preserved. Only
`openai:responses` and `openai:responses:compact` are exposed; Chat,
Claude and Gemini clients reach the provider through Aether's existing
cross-format conversion rather than new native endpoints.

Upstream Responses payloads are sanitized for what xAI actually rejects:
`previous_response_id` and `metadata.user_id` are dropped, hosted
`tool_choice` is rewritten, `web_search` is restored for converted
clients, `image_generation` is stripped on older Grok conversation
models, unsupported reasoning effort is removed, and requested
`reasoning.encrypted_content` is preserved with a replay policy keyed on
the configured provider type rather than the model name.

Quota refresh reads /user and /billing?format=credits and stores a
structured usage snapshot; a prepaid balance keeps an account selectable
after the weekly allowance is exhausted. API-key accounts skip the
subscription billing surface. The admin UI shows remaining weekly quota
as a labeled bar in the provider drawer and the pool list.

Co-Authored-By: Claude Opus 5 <[email protected]>

apps/aether-gateway/src/handlers/proxy/websocket/responses/continuation.rs

@@ -90,6 +90,8 @@ pub(super) struct ResponsesWebSocketContinuationRecord {
     /// request JSON can never set it.
     #[serde(default)]
     deepseek_opaque_reasoning_replay: bool,
+    #[serde(default)]
+    xai_encrypted_reasoning_replay: bool,
     /// A prior turn stored PII sentinels whose restore mapping exists only on
     /// the original downstream socket. Such a chain cannot safely resume on a
     /// new socket without leaking sentinels, so lookup succeeds but bootstrap
@@ -122,6 +124,10 @@ impl ResponsesWebSocketContinuationRecord {
                 normalization.reasoning_replay_policy(),
                 crate::ai_serving::OpenAiResponsesReasoningReplayPolicy::DeepSeekOpaque
             ),
+            xai_encrypted_reasoning_replay: matches!(
+                normalization.reasoning_replay_policy(),
+                crate::ai_serving::OpenAiResponsesReasoningReplayPolicy::XaiEncrypted
+            ),
             has_connection_local_redaction,
             responses_lite_static_config,
         };
@@ -156,7 +162,9 @@ impl ResponsesWebSocketContinuationRecord {
     pub(super) fn reasoning_replay_policy(
         &self,
     ) -> crate::ai_serving::OpenAiResponsesReasoningReplayPolicy {
-        if self.deepseek_opaque_reasoning_replay {
+        if self.xai_encrypted_reasoning_replay {
+            crate::ai_serving::OpenAiResponsesReasoningReplayPolicy::XaiEncrypted
+        } else if self.deepseek_opaque_reasoning_replay {
             crate::ai_serving::OpenAiResponsesReasoningReplayPolicy::DeepSeekOpaque
         } else {
             crate::ai_serving::OpenAiResponsesReasoningReplayPolicy::OpenAiItemIds
@@ -476,6 +484,7 @@ mod tests {
             binding_fingerprint: [7; 32],
             normalization_fingerprint: [9; 32],
             deepseek_opaque_reasoning_replay: false,
+            xai_encrypted_reasoning_replay: false,
             has_connection_local_redaction: false,
             responses_lite_static_config: Some(ResponsesLiteStaticConfig::from_response_create(
                 &json!({
@@ -714,6 +723,29 @@ mod tests {
         assert_eq!(decoded, record());
     }
 
+    #[test]
+    fn serialized_record_preserves_xai_replay_policy_and_reads_legacy_records() {
+        let mut expected = record();
+        expected.xai_encrypted_reasoning_replay = true;
+        let mut serialized = serde_json::to_value(&expected).unwrap();
+        let decoded: ResponsesWebSocketContinuationRecord =
+            serde_json::from_value(serialized.clone()).unwrap();
+        assert_eq!(
+            decoded.reasoning_replay_policy(),
+            crate::ai_serving::OpenAiResponsesReasoningReplayPolicy::XaiEncrypted
+        );
+        serialized
+            .as_object_mut()
+            .unwrap()
+            .remove("xai_encrypted_reasoning_replay");
+        let legacy: ResponsesWebSocketContinuationRecord =
+            serde_json::from_value(serialized).unwrap();
+        assert_eq!(
+            legacy.reasoning_replay_policy(),
+            crate::ai_serving::OpenAiResponsesReasoningReplayPolicy::OpenAiItemIds
+        );
+    }
+
     #[test]
     fn serialized_record_preserves_only_the_server_derived_reasoning_replay_policy_bit() {
         let mut expected = record();