feat(DTO): Enable codegen backend by default

[provinzkraut]

Enable the codegen backend for DTOs introduced in #2388 by default. We have initially hidden this behind a feature flag to see how it works out in production. It's been smooth sailing for about 6 months now so I think we can enable it by default. This gives us enough time to decide if we want to remove the regular backend entirely in 3.0.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.27%. Comparing base (bfeb5e9) to head (0617e10).
Report is 116 commits behind head on develop.

❗ Current head 0617e10 differs from pull request most recent head f7bccdb. Consider uploading reports for the commit f7bccdb to get more accurate results

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #3215      +/-   ##
===========================================
+ Coverage    98.23%   98.27%   +0.03%     
===========================================
  Files          312      322      +10     
  Lines        14121    14676     +555     
  Branches      2430     2330     -100     
===========================================
+ Hits         13872    14423     +551     
- Misses         107      112       +5     
+ Partials       142      141       -1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[JacobCoffee]

i wish we had some good metrics on usage here to see how many people are trying this out.

[cofin]

I agree with this approach. I've been using it for some time with no issues.

[Alc-Alc]

i wish we had some good metrics on usage here to see how many people are trying this out.

While we have two users (maintainers 🙂) using and approving this feature, I have to say I am kinda of "weary" (if this comes off as a strong word, my apologies) of this approach. I feel we should "promote" this experimental feature more. What I mean is, probably have a lot of links (backlinks?) that lead to https://docs.litestar.dev/latest/usage/dto/0-basic-use.html#enabling-the-backend.

Or even have a poll (or just a normal question) in the discord to get these two data points.

1. How many are aware and use this
2. How stable has it been

My reason for this comment is, I (not talking about other maintainers / members 🙂 ) do not think many are aware of this. How did I arrive at that conclusion? I have not seen a significant number of users ask about this in discord. To me that could mean either of the following

1. Anyone referring / getting inspired / copy pasting / carefully handpicking things from full stack have been using this and had zero issues. This is the "good ending" ✅ and we can go ahead.
2. Not many have used it and we don't have a wide spread opinion on this. This is the "bad ending" ❌ and maybe make some promotion or market this as a mandatory 3.0 alpha feature.

[peterschutt]

i wish we had some good metrics on usage here to see how many people are trying this out.

While we have two users (maintainers 🙂) using and approving this feature, I have to say I am kinda of "weary" (if this comes off as a strong word, my apologies) of this approach. I feel we should "promote" this experimental feature more. What I mean is, probably have a lot of links (backlinks?) that lead to https://docs.litestar.dev/latest/usage/dto/0-basic-use.html#enabling-the-backend.

Or even have a poll (or just a normal question) in the discord to get these two data points.

1. How many are aware and use this

2. How stable has it been

My reason for this comment is, I (not talking about other maintainers / members 🙂 ) do not think many are aware of this. How did I arrive at that conclusion? I have not seen a significant number of users ask about this in discord. To me that could mean either of the following

1. Anyone referring / getting inspired / copy pasting / carefully handpicking things from full stack have been using this and had zero issues. This is the "good ending" ✅  and we can go ahead.

2. Not many have used it and we don't have a wide spread opinion on this. This is the "bad ending" ❌ and maybe make some promotion or market this as a mandatory 3.0 alpha feature.

I see your point @Alc-Alc but the backend is supposed to be backward compatible - so as long as we do this with the guarantee that any change in behavior should be considered a bug then this shouldn't be different to rolling out any other backward compatible optimization. Also, this PR doesn't remove the current backend, so there is a pretty easy escape hatch users can apply while we address any issues.

[provinzkraut]

Definitely need to adjust the docs though, thanks for the reminder @Alc-Alc

[Alc-Alc]

I was here

[sonarcloud]

Quality Gate passed

Issues
2 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud

[github-actions]

Documentation preview will be available shortly at https://litestar-org.github.io/litestar-docs-preview/3215