v5.0.3 proposal

[rvagg]

A proposal for a patch release once #1828 is merged. This drops #1811 because it's a semver-major. What I'd like to do is rebase master on top of this once the release is out to avoid (defer?) creating a branch to manage this. It will change the hash of that commit and #1828 when it lands. If anyone objects to this then I guess we can go ahead and make a v5.x branch, I'd just prefer to not add that level of complication while we don't need it.

• [66ad305775] - python: accept Python 3 conditionally (João Reis) #1815
• [7e7fce3fed] - python: move Python detection to its own file (João Reis) #1815
• [e40c99e283] - src: implement standard.js linting (Rod Vagg) #1794
• [bb92c761a9] - test: add Node.js 6 on Windows to Travis CI (João Reis) #1812
• [7fd924079f] - test: increase tap timeout (João Reis) #1812
• [7e8127068f] - test: cover supported node versions with travis (Rod Vagg) #1809
• [24109148df] - test: downgrade to tap@^12 for continued Node 6 support (Rod Vagg) #1808
• [656117cc4a] - win: make VS path match case-insensitive (João Reis) #1806

[cclauss]

My sense is that we should have #1826 and #1827 flattened before a release.

[rvagg]

I don't think #1829 is our problem and we don't have a PR for #1826 yet. There's a little urgency to this release because we broke ecosystem things with 2761afb, see nodejs/node#28647 (comment), so I'd like to get this out sooner than later. 5.0.2 made it into an npm release unfortunately so it'd be nice to catch the next update with a new 5.0.3.

[bnoordhuis]

I merged #1828 just now.

[rvagg]

published, two commits were reordered, I'll note in their PRs what the new hashes are