Add GPT-5.1-Codex-Max model support with configurable model selection

[Copilot]

Upgrades OpenAI integration from hardcoded gpt-3.5-turbo to configurable model selection, defaulting to gpt-5.1-codex-max for enhanced code generation and analysis capabilities.

Configuration

• Added OPENAI_MODEL environment variable to .env.example (defaults to gpt-5.1-codex-max)
• Supports gpt-3.5-turbo, gpt-4, gpt-4-turbo-preview, gpt-5.1-codex-max

Backend Changes

Node.js (backend/openai.js)

// Before: hardcoded model
const completion = await openai.chat.completions.create({
  model: "gpt-3.5-turbo",
  max_tokens: 150,
  // ...
});

// After: configurable with higher token limits
const DEFAULT_MODEL = process.env.OPENAI_MODEL || 'gpt-5.1-codex-max';
const completion = await openai.chat.completions.create({
  model: model || DEFAULT_MODEL,
  max_tokens: 500,
  // ...
});

Python (ai_scripts/sample_ai.py)

• Similar pattern: environment-based model selection
• Increased token limits: 300→500 (contract analysis), 200→400 (explanations)
• Enhanced system prompts for code-focused tasks

Application

• index.js displays active model in startup banner
• Both Node.js and Python demos show model configuration

Documentation

• Added "GPT-5.1-Codex-Max Features" section detailing capabilities and use cases
• Model selection guide with comparison table
• Updated all usage examples to reflect new model

Model selection is backward compatible—users can override via OPENAI_MODEL environment variable.

Original prompt

I can’t view PR #9 in lippytm/Web3AI due to an access/tool error. Please either grant Copilot Chat access to the repo or paste the relevant files/changes here, and tell me what you want updated with GPT-5.1-Codex-Max (e.g., add env vars, config, README notes, dependency). Once I can see the PR or have the file context, I’ll add the GPT-5.1-Codex-Max updates.

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

[lippytm]

Do whatever needs to be done