fix(engine): fix broken URL resolution in Ollama and MLX engines

Both OllamaEngine and MLXEngine had two bugs in URL construction:

1. `axios.create({url: ...})` was used instead of `baseURL`, but `url`
   in axios config sets a default request URL - not a base prefix. This
   caused the URL to be ignored when `.post()` was called with a path.

2. `this.client.getUri(this.config)` was used to resolve the POST URL,
   but passing the engine config (which contains non-axios properties
   like `apiKey`, `model`, etc.) produced malformed URLs. When
   `apiKey` is null (the default for Ollama), the URL resolved to
   `http://localhost:11434/null`, returning HTTP 405.

Fix: construct the full endpoint URL once in the constructor and pass
it directly to `axios.post()`, matching how FlowiseEngine already works.

Co-Authored-By: Claude <noreply@anthropic.com>

src/engine/mlx.ts

@@ -6,16 +6,21 @@ import { AiEngine, AiEngineConfig } from './Engine';
 
 interface MLXConfig extends AiEngineConfig {}
 
+const DEFAULT_MLX_URL = 'http://localhost:8080';
+const MLX_CHAT_PATH = '/v1/chat/completions';
+
 export class MLXEngine implements AiEngine {
   config: MLXConfig;
   client: AxiosInstance;
+  private chatUrl: string;
 
   constructor(config) {
     this.config = config;
+
+    const baseUrl = config.baseURL || DEFAULT_MLX_URL;
+    this.chatUrl = `${baseUrl}${MLX_CHAT_PATH}`;
+
     this.client = axios.create({
-      url: config.baseURL
-        ? `${config.baseURL}/${config.apiKey}`
-        : 'http://localhost:8080/v1/chat/completions',
       headers: { 'Content-Type': 'application/json' }
     });
   }
@@ -31,10 +36,7 @@ export class MLXEngine implements AiEngine {
       stream: false
     };
     try {
-      const response = await this.client.post(
-        this.client.getUri(this.config),
-        params
-      );
+      const response = await this.client.post(this.chatUrl, params);
 
       const choices = response.data.choices;
       const message = choices[0].message;

src/engine/ollama.ts

@@ -6,25 +6,27 @@ import { AiEngine, AiEngineConfig } from './Engine';
 
 interface OllamaConfig extends AiEngineConfig {}
 
+const DEFAULT_OLLAMA_URL = 'http://localhost:11434';
+const OLLAMA_CHAT_PATH = '/api/chat';
+
 export class OllamaEngine implements AiEngine {
   config: OllamaConfig;
   client: AxiosInstance;
+  private chatUrl: string;
 
   constructor(config) {
     this.config = config;
 
+    const baseUrl = config.baseURL || DEFAULT_OLLAMA_URL;
+    this.chatUrl = `${baseUrl}${OLLAMA_CHAT_PATH}`;
+
     // Combine base headers with custom headers
     const headers = {
       'Content-Type': 'application/json',
       ...config.customHeaders
     };
 
-    this.client = axios.create({
-      url: config.baseURL
-        ? `${config.baseURL}/${config.apiKey}`
-        : 'http://localhost:11434/api/chat',
-      headers
-    });
+    this.client = axios.create({ headers });
   }
 
   async generateCommitMessage(
@@ -37,10 +39,7 @@ export class OllamaEngine implements AiEngine {
       stream: false
     };
     try {
-      const response = await this.client.post(
-        this.client.getUri(this.config),
-        params
-      );
+      const response = await this.client.post(this.chatUrl, params);
 
       const { message } = response.data;
       let content = message?.content;