feat: Add explicit proxy configuration support to genai.configure()

[Kartikayy007]

Description of the change

Added explicit proxy configuration support to genai.configure() with:

• HTTP/HTTPS proxy configuration via http_proxy/https_proxy parameters
• gRPC proxy support through grpc_proxy parameter
• No-proxy list specification using no_proxy
• Automatic Requests session configuration
• Dual environment variable setup (lowercase + uppercase)
• End-to-end proxy handling for all API calls

sequenceDiagram
    participant User
    participant SDK
    participant Proxy
    participant GeminiAPI
    
    User->>SDK: genai.configure(<br>http_proxy="proxy.example:8080",<br>https_proxy="proxy.example:8080")
    SDK->>Proxy: Set environment variables
    SDK->>Proxy: Configure requests.Session
    User->>SDK: generate_content()
    SDK->>Proxy: Route request through
    Proxy->>GeminiAPI: Forward API call
    GeminiAPI-->>Proxy: Response
    Proxy-->>SDK: Return result
    SDK-->>User: Final response

Loading

Motivation

• Solves connection issues in proxy-restricted environments (chat_session.send_message throw exception while using proxy config： raise ConnectionError(e, request=request) requests.exceptions.ConnectionError: HTTPSConnectionPool(host='generativelanguage.googleapis.com', port=443) #356)
• Provides first-class proxy configuration instead of relying solely on env vars
• Aligns with Python ecosystem proxy conventions
• Enables enterprise network configurations
• Supports both REST and gRPC transports

Type of change

Feature request

Checklist

• I have performed a self-review of my code
• I have added detailed comments to my code where applicable
• I have verified that my change does not break existing code
• My PR is based on the latest changes of the main branch
• I am familiar with the Google Style Guide
• I have read through the Contributing Guide

Usage Examples

Explicit configuration:

import google.generativeai as genai

genai.configure(
    api_key="YOUR_API_KEY",
    http_proxy="http://corp-proxy:8080",
    https_proxy="http://corp-proxy:8080",
    no_proxy="localhost,127.0.0.1",
    grpc_proxy="http://corp-proxy:8081"
)

model = genai.GenerativeModel('gemini-pro')
response = model.generate_content("Explain quantum physics")

Environment variable fallback:

import os
import google.generativeai as genai

os.environ["HTTP_PROXY"] = "http://corp-proxy:8080"
os.environ["HTTPS_PROXY"] = "http://corp-proxy:8080"

genai.configure(api_key="YOUR_API_KEY")  # Auto-detects proxy config

model = genai.GenerativeModel('gemini-pro')
response = model.generate_content("Explain quantum physics")

Mixed configuration:

import google.generativeai as genai

# Explicit config takes priority over env vars
genai.configure(
    https_proxy="http://alt-proxy:8080",
    api_key="YOUR_API_KEY"
)

model = genai.GenerativeModel('gemini-pro')
response = model.generate_content("Explain quantum physics") 
# Uses https_proxy="http://alt-proxy:8080" regardless of env vars

Note: Test cases will be added in a follow-up PR after initial code review.

[Kartikayy007]

@evansenter please review