fix(cli): fail fast on unknown config flags

[sanjimoh]

Previously, the OM CLI would silently ignore unknown arguments when parsing config-related flags due to the use of flags.IgnoreUnknown in loadConfigFile. This could lead to user mistakes going unnoticed.

This change removes flags.IgnoreUnknown and uses strict parsing for config flags, ensuring that any unknown or misspelled config-related arguments cause the CLI to fail fast with a helpful error message.

This improves user feedback and prevents misconfiguration due to typos or unsupported flags.

[drich10]

With the command om -e om.yml upload-product -p tiles-target/cf-6.0.5-build.2.pivotal -r 18000 -o 18000, where -r and -o are not valid command options (but are valid application options) - I expect a failure complaining one or both of those options are invalid for the command. The current changeset instead incorrectly complains that -p is an invalid option for the command.

[sanjimoh]

With the command om -e om.yml upload-product -p tiles-target/cf-6.0.5-build.2.pivotal -r 18000 -o 18000, where -r and -o are not valid command options (but are valid application options) - I expect a failure complaining one or both of those options are invalid for the command. The current changeset instead incorrectly complains that -p is an invalid option for the command.

@drich10, it was a misunderstanding on my part regarding the problem. Thank you for helping me understand the problem statement. I’ve pushed a new changeset with the intention of fixing this issue now.

$ om -e $OM_ENV upload-product -p tas_hub_collector-10.1.1747720947.pivotal -r 18000 -o 18000
Error: unknown flag(s) ["-r" "-o"] for command 'upload-product'
See 'om upload-product --help' for available options.
unknown flag(s) ["-r" "-o"] for command 'upload-product'

Could you please review this new changeset?

[drich10]

Overall it looks pretty good! Two things:

• I'd just like to see the test moved to the acceptance suite and tie the tests to some of the existing commands to tie the tests to the currently supported interface
• Can you explain the reasoning for the stderr printing on top of the error return? Its a departure from the output that occurs currently when a flag is incorrectly specified.

[sanjimoh]

Overall it looks pretty good! Two things:

• I'd just like to see the test moved to the acceptance suite and tie the tests to some of the existing commands to tie the tests to the currently supported interface
• Can you explain the reasoning for the stderr printing on top of the error return? Its a departure from the output that occurs currently when a flag is incorrectly specified.

Thanks @drich10 for the review! I've addressed both the review comments now. PTAL & let me know if it looks good to merge now.

[drich10]

Thanks for the fast changes and improved commentary on the sticky bits. Last change w.r.t logging and program control flow.

[drich10]

Thanks for the changes and looks good to me!