Improve Error Clarity on deployment fail

[jaredoconnell]

Please describe what you would like to see in this project

Currently, when the plugins are invalid, the following error is intended:

2023-10-12T18:34:17-04:00 debug source=plugin-provider Reading plugin schema...
2023-10-12T18:34:17-04:00 debug source=plugin-provider Hello message read, ATP version 2.
2023-10-12T18:34:17-04:00 error source=plugin-provider Unsupported plugin version. %!w(*errors.errorString=&{unsupported atp version '2', supported versions: [1 3]})
2023-10-12T18:34:17-04:00 error source=main Invalid workflow (invalid workflow (failed to load schema for step short_wait (failed to read plugin schema from quay.io/arcalot/arcaflow-plugin-test-impl-go:with-sdkv0.4.0-beta.1-1 (unsupported plugin version: unsupported atp version '2', supported versions: [1 3]))))

But when deployment fails, the workflow may not be the problem. The problem may be the endpoint it's connecting to, like kubernetes, docker, or podman. In which case, it may not be the workflow that's invalid. If you read deep enough down the chain of errors, this may be clear, but the leading error is "Invalid Workflow", which can be misleading.

For example, in this case, my docker or podman VM wasn't running (on mac os, since I need to start the vm for it to work):

2023-10-12T18:30:28-04:00 error source=main Invalid workflow (invalid workflow (failed to load schema for step long_wait (failed to read plugin schema from quay.io/arcalot/arcaflow-plugin-test-impl-go:with-sdkv0.4.0-beta.1-1 (failed to decode hello message (EOF)))))

It would be helpful to differentiate between deployer issues and plugin issues.