Usage reporting: fix TS declaration of fieldLevelInstrumentation

[glasser]

Note that the const fieldLevelInstrumentation which this got assigned
to actually already inferred (roughly) the correct type because the type
got merged with a number-returning function from another case.

[changeset-bot]

⚠️ No Changeset found

Latest commit: bd9e3f9

Merging this PR will not cause a version bump for any packages. If these changes should not result in a new version, you're good to go. If these changes should result in a version bump, you need to add a changeset.

Click here to learn what changesets are, and how to add one.

Click here if you're a maintainer who wants to add a changeset to this PR

[netlify]

✅ Deploy Preview for apollo-server-docs canceled.

Name	Link
🔨 Latest commit	bd9e3f9
🔍 Latest deploy log	https://app.netlify.com/sites/apollo-server-docs/deploys/62eafc17b6e37c00088d8ac3

[codesandbox-ci]

This pull request is automatically built and testable in CodeSandbox.

To see build info of the built libraries, click here or the icon next to each commit SHA.

Latest deployment of this branch, based on commit bd9e3f9:

Sandbox	Source
Apollo Server Typescript	Configuration
Apollo Server	Configuration

[trevor-scheer]

Should fieldLevelInstrumentation also be assignable to a boolean value? Not a blocker - same functionality can be achieved with 0 and 1 as it currently exists.

[glasser]

@trevor-scheer I think my reasoning here was: fieldLevelInstrumentation: 0.01 is not the same thing as fieldLevelInstrumentation: () => 0.01 (it's a shorthand for something more complex, and is what we expect many users to do), so having the number option makes sense. But fieldLevelInstrumentation: false would be identical to fieldLevelInstrumentation: () => false, so why provide two ways to say the same thing.

Arguably we should have only supported a function here, and exported a function like function weightedSample(x: number) { return async () => Math.random() < x ? 1/x : 0 } from the package instead, and the recommended usage would be fieldLevelInstrumentation: weightedSample(0.01). (Maybe if we had had deep imports in AS3 that would have felt better.)

Anyway this PR is just about fixing the types to match the implementation, not about changing the implementation.