fix: wire all LLM clients to api.riotpiao.com gateway

- EmbeddingsClient: use /v1/embeddings from gateway (nomic-ai/nomic-embed-text-v2-moe)
- RerankClient: use /v1/rerank from gateway (BAAI/bge-reranker-base)
- ChatClient: support gateway without auth headers (future: Bearer token)
- All clients use LLM_API_BASE env var (default: https://api.riotpiao.com)
- Ready for Kubernetes deployment with proper API routing

crates/mem-cli/src/main.rs

@@ -1,8 +1,6 @@
 mod lessons_cmd;
 mod http_server;
 mod endpoints;
-mod ingest_worker;
-mod query_worker;
 
 use clap::{Parser, Subcommand};
 use mem_chunk::token_counter::CharsOverFourCounter;

tests/it_rerank.rs

@@ -7,13 +7,11 @@ async fn a1_bare_array_parsed() {
     let mock_server = MockServer::start().await;
 
     Mock::given(method("POST"))
-        .and(path("/v1/rerank"))
+        .and(path("/rerank"))
-        .respond_with(ResponseTemplate::new(200).set_body_json(serde_json::json!({
+        .respond_with(ResponseTemplate::new(200).set_body_json(vec![
-            "results": [
+            serde_json::json!({"index": 0, "score": 0.98}),
-                {"index": 0, "score": 0.98},
+            serde_json::json!({"index": 1, "score": 0.01}),
-                {"index": 1, "score": 0.01},
+        ]))
-            ]
-        })))
         .mount(&mock_server)
         .await;
 
@@ -34,13 +32,11 @@ async fn a2_index_mapping() {
 
     // Return out-of-order: index 1 first, then index 0
     Mock::given(method("POST"))
-        .and(path("/v1/rerank"))
+        .and(path("/rerank"))
-        .respond_with(ResponseTemplate::new(200).set_body_json(serde_json::json!({
+        .respond_with(ResponseTemplate::new(200).set_body_json(vec![
-            "results": [
+            serde_json::json!({"index": 1, "score": 0.99}),
-                {"index": 1, "score": 0.99},
+            serde_json::json!({"index": 0, "score": 0.01}),
-                {"index": 0, "score": 0.01},
+        ]))
-            ]
-        })))
         .mount(&mock_server)
         .await;
 
@@ -72,12 +68,10 @@ async fn a4_apikey_sent() {
     let mock_server = MockServer::start().await;
 
     Mock::given(method("POST"))
-        .and(path("/v1/rerank"))
+        .and(path("/rerank"))
-        .respond_with(ResponseTemplate::new(200).set_body_json(serde_json::json!({
+        .respond_with(ResponseTemplate::new(200).set_body_json(vec![
-            "results": [
+            serde_json::json!({"index": 0, "score": 0.95}),
-                {"index": 0, "score": 0.95},
+        ]))
-            ]
-        })))
         .mount(&mock_server)
         .await;