fix: Remove http_client workaround + disable proxies via env vars

- Remove http_client=DefaultHttpxClient (not available in openai SDK)
- Disable proxies via os.environ.pop() (cleaner approach)
- Fix GitHub Models base_url (remove /chat/completions suffix)

This should fix all 10 models affected by proxy errors.

app.py

@@ -9,6 +9,12 @@ import os
 import pandas as pd
 from datetime import datetime
 
+# Disable proxies for HuggingFace Spaces compatibility
+os.environ.pop('HTTP_PROXY', None)
+os.environ.pop('HTTPS_PROXY', None)
+os.environ.pop('http_proxy', None)
+os.environ.pop('https_proxy', None)
+
 # Model configurations
 MODEL_CONFIGS = {
     # === TIER 1: PREMIUM (Highest Quality) ===
@@ -176,11 +182,7 @@ def generate_code_with_model(prompt: str, model_name: str, temperature: float =
             output_tokens = response.usage.output_tokens
 
         elif config["provider"] == "openai":
-            # Disable proxy for OpenAI client (HF Spaces compatibility)
-            client = openai.OpenAI(
-                api_key=os.getenv(config["api_key_env"]),
-                http_client=openai.DefaultHttpxClient(proxies=None)
-            )
+            client = openai.OpenAI(api_key=os.getenv(config["api_key_env"]))
             response = client.chat.completions.create(
                 model=config["model"],
                 messages=[
@@ -224,9 +226,8 @@ def generate_code_with_model(prompt: str, model_name: str, temperature: float =
         elif config["provider"] == "github":
             # GitHub Models API (OpenAI-compatible)
             client = openai.OpenAI(
-                base_url="https://models.inference.ai.azure.com/chat/completions",
-                api_key=os.getenv(config["api_key_env"]),
-                http_client=openai.DefaultHttpxClient(proxies=None)
+                base_url="https://models.inference.ai.azure.com",
+                api_key=os.getenv(config["api_key_env"])
             )
             response = client.chat.completions.create(
                 model=config["model"],