Increase interval between Heroku log drain health checks

[andygout]

The README for this repo says that the options for check objects include:

interval: time between checks in milliseconds or any string compatible with ms [default: 1minute]

We are currently experiencing the below error for several apps:

"Heroku log drain configuration check failed to fetch data: heroku responded with a 429 status code"

It seems unnecessary for this to be checked at one minute intervals because it is not a condition that will change will a high amount of regularity and given the effect it is having on contributing to reaching our Heroku API key limit, it seems judicious to reduce its impact.

Perhaps one day intervals is too extreme and happy to take a steer on what might be a more suitable interval.

[rowanmanning]

I thought we built this to only ever check when the app starts, rather than on an interval 🤔 give me a moment, I'm gonna investigate what happened here before approving (just in case we did do that and it's unclear)

[andygout]

The next-preflight check has gone from healthy to unhealthy and back again several times in the last hour, which indicates that the healthcheck is being performed with a certain amount of regularity.

If it could run only when the app starts that would definitely be the ideal (i.e. even fewer requests).

[rowanmanning]

Looks great 👍 good catch, hopefully it helps. We'll also need an n-express PR which bumps the version