Enhance transport_tracking against Error Handling for ping

[Agent-Hellboy]

currenlty we are tracking generic connection issues and retry. We Should capture specific issues much as possible as spec might suggest a different way to handle these scenarios.
e.g.
https://modelcontextprotocol.io/specification/2025-11-25/basic/utilities/ping#error-handling

• Treat ping timeouts as connection failures (log + trigger reconnect).
• Count consecutive ping failures and reset the connection after a threshold.
• Log ping failures for diagnostics.