Code review comment for lp:mir/0.18

I guess the critical question here is whether we should block release. I'm not very familiar with this part of the code, but from chatting with Andreas on IRC today, it seems like:

The linked test failures should be using the stub input platform, so they're not related to the libinput switchover.

The pointer acceleration topic (lp: #1522295) seems like something to look into, but it hasn't affected release testing, and the pointer acceleration is also something needed by PocketDesktop.

This is a judgement call (and one from someone not very entrenched in the input code at that), but it doesn't seem like this should block the release.