feat(eslint-plugin): add strict-comparisons rule

[AndreasGassmann]

Adds a strict-comparisons rule.

I also made a PR on the TSLint repo palantir/tslint#4519 which addresses an open issue palantir/tslint#3574. Once that one gets merged, I'll update the references in this PR.

I'm not sure about the name, I started calling it no-object-comparisons, but now it feels like it's doing more than just that, so strict-comparisons seems more fitting.

I'll have to migrate over some more tests from the TSLint repo, but because of the different format this takes quite some time.

[codecov]

Codecov Report

Merging #314 into master will decrease coverage by 0.09%.
The diff coverage is 92.98%.

@@            Coverage Diff            @@
##           master     #314     +/-   ##
=========================================
- Coverage    97.2%   97.11%   -0.1%     
=========================================
  Files          68       69      +1     
  Lines        2505     2562     +57     
  Branches      387      393      +6     
=========================================
+ Hits         2435     2488     +53     
- Misses         44       46      +2     
- Partials       26       28      +2

Impacted Files	Coverage Δ
...ages/eslint-plugin/src/rules/strict-comparisons.ts	92.98% <92.98%> (ø)

[bradzacher]

I can't give much of an in-depth review of the type code right now.
but test coverage and documentation looks good.

[AndreasGassmann]

So I think from my side this would be ready. Is there anything else I should do? @bradzacher @mohsen1

The corresponding tslint PR is still pending due to some compatibility issues with TS 2.1 and some new linting errors in their codebase.

[bradzacher]

a few nits, otherwise LGTM.

It would be good to see some examples/docs clarifying what cases are valid with some more types.

i.e. it's not obvious from the docs (or tests) whether I can do this without the allowObjectEqualComparison option, because your examples mostly use literals:

const x : T | null = ...;
if (x === null) { .... }

[bradzacher]

great work!

[AndreasGassmann]

@bradzacher thanks. I'll update the docs and examples to clarify the different cases, once that is done I think it's ready from my side.

awaiting docs update

[bradzacher]

update the docs and fix the merge conflict and this LGTM

[JamesHenry]

Hi @AndreasGassmann, thanks again for the PR, when do you think you will have chance to finalise this one?

[AndreasGassmann]

Sorry for not working on this for so long. I'll look at it this week.