[ViktorBystrov72]

What this PR does / why we need it:

A failed container-list or log request currently stops collection of otherwise available diagnostics. With --github-groups, a failed print command also returns before emitting the matching group end marker, so later CI output can be folded into the wrong group.

Continue collecting remaining pods and containers after individual failures, retain the printed error, and always finish each output group. When --print-logs is disabled, do not query container lists before describing the remaining pods.

Special notes for your reviewer:

The regression suite covers event failures, init/regular container discovery failures, individual log failures, disabled logs and successful runs, with GitHub grouping both on and off. It checks the collected operations, error output and balanced group markers. These failure paths reproduce on unmodified main.

Validation with Go 1.27.3 and golangci-lint 2.14.0:

go test ./... -count=1
golangci-lint run ./...
git diff --check

All passed. Tests use a fake kubectl client; no live-cluster tests were run.

[cpanato]

thanks