cedta() catches another case where utils could be top-level namespace

[MichaelChirico]

Closes #2972

Running example(":=", package="data.table", local=TRUE) winds up with ns as utils, a new case to cedta. local = FALSE has ns NULL.

Is it safe to run example in the tests? feels a bit off.

[codecov]

Codecov Report

Merging #3598 into master will increase coverage by <.01%.
The diff coverage is 83.33%.

@@            Coverage Diff             @@
##           master    #3598      +/-   ##
==========================================
+ Coverage   97.73%   97.73%   +<.01%     
==========================================
  Files          66       66              
  Lines       12910    12913       +3     
==========================================
+ Hits        12617    12620       +3     
  Misses        293      293

Impacted Files	Coverage Δ
R/cedta.R	89.47% <83.33%> (+1.97%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update c21ac65...a2fb678. Read the comment docs.

[codecov]

Codecov Report

Merging #3598 into master will increase coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master    #3598      +/-   ##
==========================================
+ Coverage   97.79%   97.81%   +0.01%     
==========================================
  Files          66       66              
  Lines       12909    12909              
==========================================
+ Hits        12625    12627       +2     
+ Misses        284      282       -2

Impacted Files	Coverage Δ
R/cedta.R	100% <100%> (+12.5%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update b25d163...c637123. Read the comment docs.