fs: move type checking for multiple fs methods to js

[jasnell]

Update errors to use internal/errors, move type checking

Checklist

• make -j4 test (UNIX), or vcbuild test (Windows) passes
• tests and/or benchmarks are included
• commit message follows commit guidelines

Affected core subsystem(s)

fs

[TimothyGu]

Code LGTM. I'd like a CITGM run though.

[maclover7]

LGTM. I'm happy to try and tackle it, but it'd be great to try and DRY up some of the validation checks, most seem somewhat similar.

[mscdex]

How does this affect performance?

[addaleax]

LGTM if benchmarks are good

[addaleax]

Something something deprecated overload ;)

[jasnell]

Looking just at fs.fstat, the impact is minimal in the microbenchmarks.

$ ./node benchmark/compare.js --old node --new ./node --filter stat fs > ~/fsbench [00:32:12|% 100| 2/2 files | 60/60 runs | 3/3 configs]: Done james@ubuntu:~/node/node$ cat ~/fsbench | Rscript benchmark/compare.R improvement confidence p.value fs/bench-stat.js statType="fstat" n=200000 -8.78 % 0.489611332 fs/bench-stat.js statType="lstat" n=200000 16.78 % * 0.039583545 fs/bench-stat.js statType="stat" n=200000 10.82 % 0.145232808 fs/bench-statSync.js statSyncType="fstatSync" n=1000000 -2.82 % ** 0.005483922 fs/bench-statSync.js statSyncType="lstatSync" n=1000000 -5.78 % ** 0.001044621 fs/bench-statSync.js statSyncType="statSync" n=1000000 -1.83 % 0.437744271 

(comparing with 9.2.0)

[jasnell]

CI: https://ci.nodejs.org/job/node-test-pull-request/11756/

[mcollina]

LGTM but it needs a CITGM run.

[jasnell]

CITGM: https://ci.nodejs.org/view/Node.js-citgm/job/citgm-smoker/1150/
New CI: https://ci.nodejs.org/job/node-test-pull-request/12087/

[jasnell]

Unrelated failures in Linux CI but running again there just to be safe: https://ci.nodejs.org/job/node-test-commit-linux/14937/

[jasnell]

Landed in 04ae486...805dca1