Release 2.30.1

[rok-cesnovar]

This .1 release only includes changes to stanc3, the cmdstan sources stay exactly the same, apart from the version change.

• Create the Stanc3 Release notes.

- Improved the pedantic mode warning for parameters without priors. The message now highlights when it is safe to ignore it.
- Fixed a bug in the memory pattern optimization where a for loop generated for the function inliner would prevent detection of for loops higher in the control flow graph.
- Fixed an issue with the function inliner when a function local variable had the same name as a variable in the calling scope.
- Fixed an issue where recursive functions which called themselves with a slice of their arguments would lead to infinite template expansion in C++ compilation.
- Fixed a bug in stanc.js where the stanc flags were not listed correctly in the generated C++.
- The imaginary components of complex vectors and matrices are now initialized to NaN to match the real component and behaviour of non-complex vectors and matrices.

• Update the version numbers on CmdStan master branch.
• Create the Stanc3 release.
• Create x86 CmdStan tarballs (check version, check that the extracted folder is in the cmdstan-version format).
• Make sure the CmdStan sources in the tarball have not changed from the 2.30.0 release.
• Create non-x86 CmdStan tarballs.
• Create conda CmdStan tarballs.
• Make a Stan blog release announcement post.
• Post an update in the 2.30 release announcement Discourse thread
• Make a Twitter announcement.

[serban-nicusor-toptal]

Hey, everything should be done now.
FYI we did 2.30.0 from the develop branch last time, so I went in and created a branch v2.30.0 out of it, then branched off that branch to create 2.30.1 + version update
( I will create branch 4.4.0/2.30.0 for math and stan now too ) ( done )
If someone can double check if everything is fine that would be awesome, thanks!

And as always, thanks @rok-cesnovar for these awesome issues that are super helpful every time!

[rok-cesnovar]

Thanks! The tarballs look good at first glance. The stanc3 binaries return the correct version and the cmdstan tarball has the correct version. Running cmdstanr tests now to double-check.

Let's make sure we merge the 2.30.1 branch to the master cmdstan branch once this is confirmed to be good (same for Stan and Math (4.4.0 there)).

[serban-nicusor-toptal]

As soon as this is done here, I will create PRs and post them here for approval. Thank you for double-checking!

[WardBrian]

Thanks @serban-nicusor-toptal and @rok-cesnovar! I'm also running the cmdstanpy tests here https://github.com/stan-dev/cmdstanpy/actions/runs/2748213992 but don't expect anything interesting

[rok-cesnovar]

Tests passed in cmdstanr as well. Tweeted the announcement as well: https://twitter.com/mcmc_stan/status/1552368360670855172

@serban-nicusor-toptal once you merge everything to master, feel free to close this issue. Thanks!

[serban-nicusor-toptal]

PR for stan: stan-dev/stan#3133
PR for math: stan-dev/math#2799

Seems there are a lot of changes and a few conflicts

[WardBrian]

For stanc I think we just need a PR to update the release notes

[serban-nicusor-toptal]

For stan, just the release notes + versions: stan-dev/stan#3134
I see that math is failing, it's trying to run downstream_hotfix (from stan master branch) and failing with

make doxygen
/bin/sh: null: command not found
lib/stan_math/make/libraries:117: *** "Need to set TBB_CXX_TYPE for non-standard compiler other than gcc or clang.".  Stop.
script returned exit code 2

I'm not entirely sure why, I assume because of some discrepancy in code/Jenkinsfile. Does anyone know more about this?
PS: The PR just created is also failing with the same error ( branched-off from master )

[serban-nicusor-toptal]

I've created a few new PRs:

• merge 2.30.1 release into cmdstan develop: Update version and release notes in develop #1111 ( will need to create a new one after to merge develop -> master )

[serban-nicusor-toptal]

Please find stanc3 PR here: stan-dev/stanc3#1235

[serban-nicusor-toptal]

PR for merging v2.30.1 cmdstan into master: #1112

[WardBrian]

It seems in this release the minor version is not set properly in compiled executables:

(cmdstan) [brian@FlatTop cmdstan] $ ./bin/stanc --version
stanc3 v2.30.1 (Unix)
(cmdstan) [brian@FlatTop cmdstan] $ ./examples/bernoulli/bernoulli info
stan_version_major = 2
stan_version_minor = 30
stan_version_patch = 0

This also impacts the output.csv version reported.

I don't think this is a critical issue to fix, but it might be important to understand why it occurred