http: remove unused hasItems() from freelist

[Trott]

Remove the hasItems() method from freelist module as it is unused
internally.

Checklist

• make -j4 test (UNIX), or vcbuild test (Windows) passes
• commit message follows commit guidelines

[Trott]

Marking semver-major out of an abundance of caution as the API is exposed in one place:

$ node -p "require('_http_common').parsers.hasItems()"
false
$

[Trott]

CITGM: https://ci.nodejs.org/view/Node.js-citgm/job/citgm-smoker/2099/

[nodejs-github-bot]

CI: https://ci.nodejs.org/job/node-test-pull-request/27241/ ✔️

[Trott]

CITGM on master for comparison: https://ci.nodejs.org/view/Node.js-citgm/job/citgm-smoker/2104/

(I'm seeing more commander failures than I expected on the CITGM for this PR. But that seems unrelated. So, running against master to see how it does there.)

[Trott]

And...CITGM this PR: https://ci.nodejs.org/view/Node.js-citgm/job/citgm-smoker/2105/

[Trott]

The commander results for this PR continue to be puzzling to me. One more CITGM against master to see if it shows up there or not: https://ci.nodejs.org/view/All/job/citgm-smoker/2108/

[BridgeAR]

@Trott the commander results are also on v13, so it's definitely independent from this PR.

[Trott]

@Trott the commander results are also on v13, so it's definitely independent from this PR.

Not just the fact that commander is failing, but that it fails the way it does in this PR?

On master, commander fails in two or three places for a few different reasons. Here, it fails in six places each time, and every time it's the same way, with a 5000ms timeout being exceeded (probably due to either a race condition or an async event not running at all).

[Trott]

@Trott the commander results are also on v13, so it's definitely independent from this PR.

Not just the fact that commander is failing, but that it fails the way it does in this PR?

On master, commander fails in two or three places for a few different reasons. Here, it fails in six places each time, and every time it's the same way, with a 5000ms timeout being exceeded (probably due to either a race condition or an async event not running at all).

Just checked for myself, and yeah, they show up on 13.2.0 CITGM, but a little less frequently. They've have to be unrelated. I'm being super-cautious though. Sorry for all the notifications.

[Trott]

One hopefully-last CITGM against this PR: https://ci.nodejs.org/job/citgm-smoker/2109/

[Trott]

OK, the CITGM commander issue showed up on master too so I think this is good to land. I'm still going to let this one last CITGM run. I'm still bugged that it seems to occur about twice as often with this change, which again, makes no sense, but stranger stuff has happened....

[Trott]

(Also, it's happening but only very infrequently for me locally on master.)

[Trott]

OK, the CITGM commander issue showed up on master too so I think this is good to land. I'm still going to let this one last CITGM run. I'm still bugged that it seems to occur about twice as often with this change, which again, makes no sense, but stranger stuff has happened....

Last CITGM run with this PR, the timeout happened just twice, totally in line with master branch CITGM. Landing!

[Trott]

Landed in ff2ed3e