feat: Update Emulator configuration function

[ajupazhamayil]

Resolves #717

BEGIN_COMMIT_OVERRIDE
feat: add emulator support for Bigtable and Spanner clients (#719)
END_COMMIT_OVERRIDE

[ajupazhamayil]

Investigating why the code changes didn't update the integration tests goldens!!

[ajupazhamayil]

Integration test files are manually updated, introduced some errors to see if the cache delete takes effect. Apparently, the integration tests are running irrespective of the code changes!!

PS: Command used to rotate the cache key: uuidgen | gh secret set CACHE_VERSION -r https://github.com/googleapis/gapic-generator-php

cc: @bshaffer

[bshaffer]

This looks great! Only suggestion is we should use the noal-coalescing assignment operator (??=) to ensure that these are defaults that can be overwritten by the user.

See #720 where I've added the operator to the AST

[bshaffer]

The only problem I see is that where before this was providing default values, it's now overwriting any user-supplied values. What if we use ??= instead?

Suggested change

	$options['apiEndpoint'] = $emulatorHost;
	$options['transportConfig']['grpc']['stubOpts']['credentials'] = ChannelCredentials::createInsecure();
	$options['credentials'] = new InsecureCredentialsWrapper();
	$options['apiEndpoint'] ??= $emulatorHost;
	$options['transportConfig']['grpc']['stubOpts']['credentials'] ??= ChannelCredentials::createInsecure();
	$options['credentials'] ??= new InsecureCredentialsWrapper();

[ajupazhamayil]

Once the customer set the emulator environment variable, endpoint and credentials should be REPLACED to the emulator end point. WDYT?

Ref: https://github.com/googleapis/google-cloud-php/blob/400f247922efc876582a5fb39cf8072e0df63420/Spanner/src/Connection/Grpc.php#L288

[bshaffer]

I personally don't like that, I think they should just be defaults... I don't ever think an environment variable should REPLACE something in code - they should always function as a default