Creating a DevWorkspace with memoryRequest > memoryLimit incorrectly results in WorkspaceEngineFailure

[dkwon17]

Describe the bug

When deploying a DevWorkspace with a container where memoryRequest is greater than memoryLimit results a WorkspaceEngineFailure instead of a BadRequestFailure.

Che version

other (please specify in additional context)

Steps to reproduce

1. Checkout feat: Add WorkspaceEngineFailure reason to the devworkspace_fai… devfile/devworkspace-operator#647 (DevWorkspace startup failure reasons are set in this PR), and follow instructions from the PR to run locally.
2. Create a DevWorkspace with this YAML file in the devworkspace-controller namespace: https://gist.github.com/dkwon17/13c85b162874b9bf34abc50d576aa494
3. When the DevWorkspace startup fails, view the DevWorkspace's conditions. The FailedStart condition's reason field is WorkspaceEngineFailure (see screenshot above)

Expected behavior

The reason field should be BadRequest, since a memoryRequest value greater than the memoryLimit is invalid.

Runtime

CodeReady Container

Screenshots

No response

Installation method

other (please specify in additional context)

Environment

macOS

Eclipse Che Logs

No response

Additional context

The FailedStart condition's reason field is set thanks to this PR: devfile/devworkspace-operator#647.

This issue was created to be a followup for the work done that PR, to improve the way failure reasons are determined.

[amisevsk]

This issue would be better suited for the devworkspace-operator repo. I think the ultimate fix may belong in the (further) upstream devfile/api repo, as they provide the basic validation package used in DWO for validating DevWorkspaces.

[amisevsk]

devfile/api#672 tracks this as a validation item in devfile/api.