Work around ingress glitch with 304 responses

[masto]

Proposed change

This is a small workaround for an issue I noticed when using the ESPHome add-on (though it is not specific to that add-on). The ingress proxy rewrites "304 Not Modified" responses into chunked transfer-encoding, which triggers a series of weird browser behaviors. I don't know exactly why, but I wrote up the observed symptoms and in detail in esphome/issues#2908. This one-line change makes the problem go away.

Type of change

• Dependency upgrade
• Bugfix (non-breaking change which fixes an issue)
• New integration (thank you!)
• New feature (which adds functionality to an existing integration)
• Breaking change (fix/feature causing existing functionality to break)
• Code quality improvements to existing code or addition of tests

Additional information

• This PR fixes or closes issue: fixes "An error occurred saving" message in Safari due to bad ingress proxying for 304 responses esphome/issues#2908
• This PR is related to issue:
• Link to documentation pull request:

Checklist

• The code change is tested and works locally.
• Local tests pass. Your PR cannot be merged unless tests pass
• There is no commented out code in this PR.
• I have followed the development checklist
• The code has been formatted using Black (black --fast homeassistant tests)
• Tests have been added to verify that the new code works.

If user exposed functionality or configuration variables are added/changed:

• Documentation added/updated for www.home-assistant.io

If the code communicates with devices, web services, or third-party tools:

• The manifest file has all fields filled out correctly.
  Updated and included derived files by running: python3 -m script.hassfest.
• New or updated dependencies have been added to requirements_all.txt.
  Updated by running python3 -m script.gen_requirements_all.
• For the updated dependencies - a link to the changelog, or at minimum a diff between library versions is added to the PR description.
• Untested files have been added to .coveragerc.

The integration reached or maintains the following Integration Quality Scale:

• No score or internal
• 🥈 Silver
• 🥇 Gold
• 🏆 Platinum

To help with the load of incoming pull requests:

• I have reviewed two other open pull requests in this repository.

[probot-home-assistant]

Hey there @home-assistant/supervisor, mind taking a look at this pull request as it has been labeled with an integration (hassio) you are listed as a code owner for? Thanks!
_{^{(message by CodeOwnersMention)}}

[elupus]

It seems a zero length chunk is indicative of end of stream. So it's expected to close the connection after this has been received. The client should re-open a new connection on the next request. I wonder what is getting confused by this closed connection.

[elupus]

I think this is correct, we should however also add 204 "no content" responses. Both 304 and 204 are expected empty responses, that will "break" chunked encoding.

[masto]

As you say, closing the connection may well be correct behavior. (I got so caught up in the weeds I probably lost sight of that). Why it breaks Safari is a mystery -- perhaps there's some tiny out-of-spec implementation detail that can be found with more detailed inspection. All I can say is that Safari does something weird under these conditions.

[frenck]

Tagged it for the milestone (although... it does affect Ingress, so we should be careful)

[ludeeus]

This should use (HTTPStatus.NO_CONTENT, HTTPStatus.NOT_MODIFIED) and should have a test that proves that these 2 statuses return Response and not StreamResponse