feat: warn when clang-format fails

[kjin]

No description provided.

[ofrobots]

Odd that travis wasn't launched for this.

[kjin]

@ofrobots Probably because it's unmergeable... rebased

[DominicKramer]

LGTM: provided the tests pass

[ofrobots]

LGTM if the coverage impact is positive.

[codecov-io]

Codecov Report

Merging #62 into master will increase coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master      #62      +/-   ##
==========================================
+ Coverage   99.58%   99.59%   +0.01%     
==========================================
  Files           8        8              
  Lines         241      249       +8     
  Branches        8       10       +2     
==========================================
+ Hits          240      248       +8     
  Misses          1        1

Impacted Files	Coverage Δ
test/test-kitchen.ts	100% <100%> (ø)	⬆️
src/clean.ts	100% <100%> (ø)	⬆️

---

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 0f8a049...dbb6630. Read the comment docs.

[ofrobots]

When landing, make sure the commit includes the 'Fixes: #60' metadata.

[kjin]

Since the approvals were issued, I added a test... was able to get code coverage to increase by 0.01%

A point of possible contention is here. The explanation is that exec shouldn't actually be promisified the way it is because the callback may populate err and stdout/stderr. So I added a new function to always resolve with an exit code (though most places still use the now-renamed "simple" execp)

Also, two things happened:

• clang-format updated, meaning things had to be re-formatted
• chalk got built-in TS definitions that clash with @types/chalk, so I removed the dependency on the latter (2e4711a)

@DominicKramer @ofrobots PTAL at the two linked things.

[ofrobots]

This deserves a comment. Also, we should probably look for an ecosystem module for execp.

[ofrobots]

Rule of thumb: if you have explain why some code is the way it is in the PR, you probably also need it as a comment.

LGTM once that is addressed.