Fix/prometheusexporter Shutdown HTTP server

[Argannor]

Description: Shutdown the http.Server instance on exporter shutdown

Link to tracking Issue: #35464

Testing: Manual testing. I included this version of the exporter in our internal distribution and deployed it to verify the metrics no longer go stale after receiving a NOHUP signal.

Documentation: n/a

[linux-foundation-easycla]

The committers listed above are authorized under a signed CLA.

• ✅ login: Argannor / name: Argannor (52eee00, 40d9f16, 3a017df, c59034c, 6efa063, 7b7535f)
• ✅ login: dashpole / name: David Ashpole (925a967)

[Argannor]

Hey @Aneurysm9, @songy23, sorry for pinging you directly. Do you mind having a look at this or advising me what is needed from me to go ahead with this PR?

[dashpole]

unit tests seem to fail after the change.

[Argannor]

Yeah, I just took a look at the teste that are failing, and I'm not sure what's happening there:

	t.Cleanup(func() {
		require.NoError(t, exp.Shutdown(context.Background()))
		// trigger a get so that the server cleans up our keepalive socket
		var resp *http.Response
		resp, err = httpClient.Get("https://localhost:7777/metrics")
		require.NoError(t, err)
		require.NoError(t, resp.Body.Close())
	})

Is this a workaround for the fact that the server did not shutdown correctly previously, and is now failing, because after the shutdown the server is shutdown?

[dashpole]

It might have been. Try removing the httpClient.Get call from cleanup and see if tests pass.

[Argannor]

After removing the statements the tests passed locally for me. Sorry for not running them in the first place, I had some troubles running them using make... :/

[Argannor]

Is there something you need me to do?