LightFilterVisualiser : Fix Arnold light blockers

[ericmehl]

This fixes a regression in #5421 that prevented Arnold Light blockers from being visualized. I had originally tested to see if the lightShaderNetwork was present and skipped the visualizer if it wasn't. Arnold light blockers don't need a lightShaderNetwork, and I don't suppose would make sense for them to have one (?) so my skip criteria was too aggressive.

I think I'm happy with this solution as-is, but is it worth not returning an empty visualiser set from visualise() and throwing an exception or something else instead?

Checklist

• I have read the contribution guidelines.
• I have updated the documentation, if applicable.
• I have tested my change(s) in the test suite, and added new test cases where necessary.
• My code follows the Gaffer project's prevailing coding style and conventions.

[johnhaddon]

LGTM. Sorry, I should have spotted this in the initial PR.

is it worth not returning an empty visualiser set from visualise() and throwing an exception or something else instead?

I think an exception would be going too far - the Viewer would stop showing anything at all then. There might be a case for emitting a warning, but I'm not sure how much value it would have. Any thoughts @murraystevenson?

[murraystevenson]

Any thoughts?

Think I'd be happy to keep the silent approach. There are cases where a warning could be annoying such as when using attribute inheritance to assign the same gobo to multiple spot lights in a group by assigning the gobo to the group itself.