Update job log output limit exceeded warning

Job logs in GitLab CI/CD have a limited length. It's designed for operational reasons, as the jobs can create arbitrary outputs so one could - by mistake or by purpose - output even gigabytes or terabytes of text.

The log limit is enforced at the Runner level. By default it's set to 4MiB and the Runner administrator can this overwrite in config.toml by using the output_limit setting at the [[runners]] section.

When the job log limit is exceeded, Runner stops collecting new output, but keeps executing the job. The job may succeed, may fail or may be canceled after the log exceeded the limit and the final job status will be still discovered properly. However, to make it clear for the user why there is no new output, a warning is printed:

The UX of this warning is however confusing:

1. Is printed in red, and GitLab CI/CD uses red in the job output to mark the errors. And errors are usually considered as something that "failed the job". While a case when job log exceed the output limit is mostly not expected, it doesn't mean it's definitely an error. However, it's often confusing people, especially when the limit was exceeded and the job failed. Users often point the log limiting as the reason of job failure, while the failure is totally not caused by that. But it's hidden behind a "not collected" output.

2. Not every user is aware that job execution is still happening after the job log limit is exceeded. Writing it clear in the job log would make it easier to share this knowledge.

Proposal

Update the UX of the log limit exceeded warning to something that will be not confused with job execution errors and will make it clear that job is still being executed.