fix: respect existing Cognito configuration during redeploy

[amitkumawat-node]

Summary

This PR fixes an issue where user-provided Cognito configuration is ignored during redeployments, causing a new User Pool to be created instead of reusing an existing one.

Problem

On initial deployment, a new Cognito User Pool is created as expected when no environment variables are provided.

However, on subsequent deployments, even when explicitly setting:

• USE_EXISTING_COGNITO=true
• EXISTING_COGNITO_USER_POOL_ID=

the script overrides this configuration based on stack detection logic and sets USE_EXISTING_COGNITO=false. As a result, a new Cognito User Pool is created on every run.

Root Cause

The deployment script prioritizes auto-detection logic over user-provided environment variables, leading to unintended overrides.

Changes

• Updated logic to check for user-provided configuration first
• Use existing Cognito User Pool when both:
  • USE_EXISTING_COGNITO=true
  • EXISTING_COGNITO_USER_POOL_ID is provided
• Fallback to existing auto-detection logic when configuration is not provided
• Preserved existing behavior for first-time deployments

Behavior After Fix

• First deployment (no env variables): creates a new User Pool
• Subsequent deployment with env variables: reuses existing User Pool
• Default auto-detection flow remains unchanged

Impact

• Prevents unnecessary creation of multiple Cognito User Pools
• Ensures idempotent deployments
• Improves predictability and user control

Testing

• Verified initial deployment creates a new Cognito User Pool
• Verified redeployment with env variables reuses the existing pool
• Verified fallback behavior when env variables are not provided

Related Issue

Fixes #41