fix: handle fractional "check result" logs

[nadiamoe]

What this PR does / why we need it:

When grafana/xk6-sm#34 is merged, it will be possible for check result logs to have fractional values, if the same check name is used twice. For example, the following snippet:

  check(res, {
      'is 200': () => true,
    }
  );
  check(res, {
      'is 200': () => false,
    }
  );

Currently yields:

time="2025-01-14T17:32:54+01:00" level=info msg="check result" check="is 200" metric=checks_total scenario=default source=sm value=1
time="2025-01-14T17:32:55+01:00" level=info msg="check result" check="is 200" metric=checks_total scenario=default source=sm value=0

After grafana/xk6-sm#34 is merged, it will yield

time="2025-01-14T17:32:54+01:00" level=info msg="check result" check="is 200" metric=checks_total scenario=default source=sm value=0.5

In terms of reporting, and given this will only happen if two checks with the same name are defined, I think it is okay to simplify and assume that if not 100% of the checks with the same name are okay, the check has failed. So this PR changes the query to look for != 1 instead of = 0. Other queries already look for = 1, so those shouldn't need changing.

[VikaCep]

In the example you provided, then with these changes both checks would fail, is that right?

[nadiamoe]

Yes, I think this would be a good approximation to handle this corner case: If at least one instance of check of a given name fails, then the check should be reported as failed.

[VikaCep]

LGTM 👍