Fix bug to handle duplicate subnet across different vpc networks

[jimmycgz]

SUMMARY

GCP doesn't allow to create any subnet with the same name in the same region if already exists in any other VPC network, but the original ansible module gcp_compute_subnetwork doesn't throw an error. So this PR suggests adding a condition check in the module and error out for this scenario.

Fixes #560

ISSUE TYPE

• Bugfix Pull Request

COMPONENT NAME

plugins/modules/gcp_compute_subnetwork.py

ADDITIONAL INFORMATION

GCP API will throw an error if found a duplicated subnet name on another vpc when you are trying to create a same-name subnet. Refer the API guide for details. https://cloud.google.com/compute/docs/reference/rest/v1/subnetworks/insert
This PR suggests adding a condition check to error out the above-mentioned scenario, by comparing the intended vpc network in the request vs the vpc found for the existing subnet, and throw an error if it's a different vpc network.

After this suggested change, A fatal error will occur to handle the above-mentioned case. Which tells the user to either change the subnet name to unique or change to a different region.

[toumorokoshi]

thanks! there's linting errors which is the main blocker for me.

Otherwise a nit that would be nice to fix, but not a blocker.

[jimmycgz]

Thanks @toumorokoshi for the suggestion. It looked good to me so I committed.