feat: add MCP trust verification example with AgentOps monitoring

[twzrd-sol]

Summary

Adds examples/mcp/ — an example demonstrating AgentOps monitoring of an MCP-powered Solana agent trust verification workflow.

What it shows:

• Connects to the TWZRD Agent Intel MCP server (streamable-HTTP) to call score_agent and preflight_check on a Solana wallet
• Gates downstream OpenAI requests behind the trust check: wallets with score ≥ 0.5 and APPROVE preflight proceed; others are rejected with a reason
• Records every step as an AgentOps trace — trust_verification action, OpenAI response, and end state (Success/Rejected) — all tagged with mcp, twzrd, solana

Why it's useful:

• Shows agentops monitoring a real external MCP tool call flow (not just OpenAI)
• Demonstrates trust-gating patterns: only verified on-chain agents can trigger LLM calls
• x402 (HTTP 402) paid endpoints (get_trust_receipt) can be metered via AgentOps for cost tracking

Files added:

• examples/mcp/mcp_trust_verification.py — full working Python script
• examples/mcp/README.md — setup, usage, and dashboard instructions
• examples/mcp/requirements.txt — dependencies

MCP server details:

• Endpoint: https://intel.twzrd.xyz/mcp (streamable-HTTP, no API key needed for free tools)
• Free tools: score_agent(wallet), preflight_check(wallet)
• Paid tool: get_trust_receipt(wallet) — HTTP 402 / x402 micropayment

[twzrd-sol]

Apologies - this PR was opened by a misconfigured autonomous agent that over-generalized its submission target list. Withdrawing it. Sorry for the noise.