Fix false positives for no-descending-specificity and scss syntax

[YozhikM]

Which issue, if any, is this issue related to?

Closes #3417

Is there anything in the PR that needs further explanation?

No.

[jeddy3]

@YozhikM Thanks. Looking good. I've made two minor requests around conventions.

[jeddy3]

As per the conventions:

Use the minimum amount of code possible to communicate the pattern e.g. if the rule targets selectors then use an empty rule e.g. {}.

Can we use the following please?:

.foo { &--a, &--a#{&}--b { div {} } }

[jeddy3]

The convention, in this file and others, is to return early when using the isStandard* utils. We should probably follow it here too:

parseSelector(resolvedSelector, result, rule, s => {
  if (isStandardSyntaxSelector(resolvedSelector)) {
    return;
  }
  checkSelector(s, rule, index, comparisonContext);
}

[jeddy3]

LGTM

[hudochenkov]

LGTM!

[hudochenkov]

• Fixed: no-descending-specificity false positives for #{&} (#3420).