Performance enhancement snippet (pre PR discussion)

The following is a working (but with possible regressions) snippet to reduce the query load of fetching many acf fields.

# Summary

The overarching principal is:

1. pass in an array of key/value pairs (the field is key, as the type is not likely to be unique)
2. a new method ("fields" or "bulk" or whatever) makes a generic `select * from `wp_postmeta` where `post_id` in ('106')` via `$this->post->meta`
3. we filter down the meta values to just the keys we passed in
4. feed those meta models back into the field factory.

This, from elementary testing, changes the query to 1 + N (1 for post + N per acf field) to 2 (1 for post + 1 for all meta).

## Before:

## After:

## Code changes

// Sample usage
Staff::find(1);
$fields = $staff->acf->fields([
'staff_short_description' => 'wysiwyg',
'staff_description' => 'wysiwyg',
'staff_role' => 'text',
]);

dd($fields); // (see screenshot)

The following ACF fields would need to be modified:

// AdvancedCustomFields.php
// Add new bulk/fields/etc method
public function fields($fields)
{
// whereIn limits the fields down to just the keys we are looking for
// feed the same stuff into the Factory
return $this->post
->meta
->whereIn('meta_key', array_keys($fields))
->mapWithKeys(function($field) use ($fields)
{
$value = FieldFactory::make($field, $this->post, snake_case($fields[$field->meta_key]))->get();
return [$field->meta_key => $value];
});
}

And BasicField (at minimum):

// BasicField
// This bit is the tricky bit - we need to remove the query from fetchValue
// possibly into a seperate method (maybe performQuery)
public function fetchValue($field)
{
// $postMeta = $this->postMeta->where(
// $this->getKeyName(), $this->post->getKey()
// )->where('meta_key', $field)->first();

if (isset($postMeta->meta_value) and ! is_null($postMeta->meta_value)) {
$value = $postMeta->meta_value;
...
// remaining code is the same

## Concerns

My major concern is I am missing something *major* - as to _why_ the per field query is needed (the removed lines in BasicField).

Obviously regression is reasonably big too.

Thoughts before kicking off a PR? I'm extremely short on time (who isn't?) at the moment - so need to be careful how I allocate resources.