newline-per-chained-call: bad indentation, and bare fn call treated as chain

[OJFord]

Validations

• Follow our Code of Conduct
• Read the Contributing Guide.
• Check that there isn't already an issue that reports the same bug to avoid creating a duplicate.

Describe the bug

With "@stylistic/newline-per-chained-call": ["error", {ignoreChainWithDepth: 1}]:

export const post = sqliteTable(
  'post',
  {
    id: integer('id').primaryKey(),
    name: text('name'),
  },
)

is reformatted to:

export const post = sqliteTable(
  'post',
  {
    id: integer('id')
.primaryKey(),
    name: text('name'),
  },
)

Which:

1. I would not expect anyway, since integer('id').primaryKey() is to me a chain of 1 - it would be fine if it were integer.primaryKey(), I expected that a 'chain' refers to occurrences of the . operator;
2. is weird indentation, not that there's really a great solution anyway

A workaround is to introduce parens:

export const post = sqliteTable(
  'post',
  {
    id: (
        integer('id')
        .primaryKey()
    ),
    name: text('name'),
  },
)

which is a bit verbose but less odd looking than the zero indentation above.

---

To address (1), I think this should not be treated as a >1 chain.

---

Assuming it were integer('id').something.primaryKey() then, to address (2), I would suggest a configurable option for what happens in object values.

/* ["error", {ignoreChainWithDepth: 1, ignoreChainInObjectValue: true}]] */

// GOOD
{
  id: foo().bar().baz(),
}

(
  foo()
  .bar()
  .baz()
)

// BAD
(
  foo().bar().baz()
)

then if it really does get too long and we want to force it, we can always:

// GOOD
{
  id: (
    foo()
    .bar()
    .baz()
  ),
}

or this could even just be new, unconfigurable, behaviour - since I'm really not sure what a sensible indentation would look like, all possibilities seem a bit weird:

{
  a: foo()
.bar()
.baz(),

  b: foo()
     .bar()
     .baz(),

  c: foo()
  .bar()
  .baz(),
}

or whatever else.

Reproduction

See description, I can't work out how to use Stackblitz for it

Contributes

• I am willing to submit a PR to fix this issue
• I am willing to submit a PR with failing tests