feat: add waitForAgentReady helper for agent deployment polling

[sathwik-14]

The helper does interval based polling for checking agent rediness.

This is needed to avoid running tight loops & respect API rate limits

Example

const agent = await waitForAgentReady(client, 'agent-123', {
timeout: 60000, // 1 minute
onUpdate: (status, elapsed) => {
console.log(`Status: ${status} (${elapsed}ms elapsed)`);
}
});

Edge Cases Handled

• Graceful Cancellation: Supports AbortSignal for canceling long-running polls
• Timeout Protection: Prevents infinite polling with configurable timeout (default: 3 minutes), checking before each API call to respect time bounds
• Deployment Failures: Detects terminal failure states (STATUS_FAILED, STATUS_UNDEPLOYMENT_FAILED) and throws structured errors
• Transitional States: Continues polling through intermediate deployment states without false positives
• Network Errors: API failures propagate to caller, allowing custom retry/error handling strategies
• Structured Error Types: Custom AgentTimeoutError and AgentDeploymentError classes enable type-safe error handling with accessible metadata
• Concurrent Polling: Stateless implementation supports multiple independent polls of the same agent
• Progress Monitoring: Optional onUpdate callback provides real-time status updates for UI/logging integration

[bbatha]

This is a solid implementation. However please implement it closer to the style of this repo. The #8 is a good reference.

That means that this should be a method on the Agent client class so that it can be used in a fluent style.

[bbatha]

Drop this

[bbatha]

Include the request options

[bbatha]

This should already be defined in the agent file

[bbatha]

This looks great can you just make sure lints and tests pass then I'll merge it

[bbatha]

Resolves #4