fix standalone generated quantities csv writer

[avehtari]

Fixes #3403

The fix and PR text were assisted by Claude

Submission Checklist

• Run unit tests: ./runTests.py src/test/unit
• Run cpplint: make cpplint
• Declare copyright holder and open-source license: see below

Summary

gq_writer::write_gq_values has two overloads; the
std::vector one drops the first num_constrained_params_ values
before writing, the Eigen one wrote everything write_array returned.
The multi-chain path of standalone_generate uses the Eigen overload,
so generate_quantities num_chains=N produced rows with the constrained
parameters prepended and a header listing only the generated quantities.

Intended Effect

Multi-chain standalone generate_quantities writes one value per header
name, matching the single-chain path.

How to Verify

./runTests.py src/test/unit/services/sample/standalone_gqs_parallel_test.cpp

New test genDraws_bernoulli_multi_chain_column_count: for a two-chain
run, every row must have as many comma-separated fields as the header
has names. Fails on develop, passes with this PR.

Command line:

./model generate_quantities fitted_params=fit.csv num_chains=2 output file=gq.csv

gq_1.csv rows go from theta_value, u_value to u_value.

One further test is not included here. Once #3402 is
merged I will rebase and add
genDraws_bernoulli_multi_chain_matches_single, which asserts that
chain i of a multi-chain run reproduces a single-chain run with chain
id 1 + i. It needs both #3402 and this PR for the column
alignment.

Side Effects

Output change: generate_quantities num_chains=N loses the spurious
leading parameter columns. Anyone whose reader adapted to the current
(header-inconsistent) layout will need to drop the workaround.

Not addressed here: the same two paths disagree on the CSV separator
(", " from the Eigen writer, "," from the other). Cosmetic, and
fixing it means touching the writer classes, which are shared with other
methods.

Documentation

None.

Copyright and Licensing

Aki Vehtari

By submitting this pull request, the copyright holder is agreeing to license the submitted work under the following licenses:

• Code: BSD 3-clause (https://opensource.org/licenses/BSD-3-Clause)
• Documentation: CC-BY 4.0 (https://creativecommons.org/licenses/by/4.0/)

[WardBrian]

Besides fixing the merge conflict, mind also adding a test that you can get the exact same result from both overloads now?

[avehtari]

That test was already waiting and now included

[stan-buildbot]

---

Name	Old Result	New Result	Ratio	Performance change( 1 - new / old )
gp_regr/gp_regr.stan	0.12	0.11	1.04	3.65% faster
gp_regr/gen_gp_data.stan	0.03	0.03	1.02	2.13% faster
arK/arK.stan	1.66	1.66	1.0	0.12% faster
eight_schools/eight_schools.stan	0.06	0.06	1.01	1.25% faster
low_dim_gauss_mix_collapse/low_dim_gauss_mix_collapse.stan	8.91	8.89	1.0	0.24% faster
pkpd/one_comp_mm_elim_abs.stan	20.52	20.45	1.0	0.37% faster
pkpd/sim_one_comp_mm_elim_abs.stan	0.26	0.27	0.99	-0.65% slower
sir/sir.stan	79.49	74.77	1.06	5.94% faster
gp_pois_regr/gp_pois_regr.stan	2.52	2.63	0.96	-4.44% slower
low_dim_gauss_mix/low_dim_gauss_mix.stan	2.96	3.1	0.95	-4.93% slower
irt_2pl/irt_2pl.stan	5.56	5.76	0.96	-3.77% slower
arma/arma.stan	0.35	0.32	1.08	7.51% faster
garch/garch.stan	0.53	0.55	0.96	-4.12% slower
low_dim_corr_gauss/low_dim_corr_gauss.stan	0.01	0.01	0.97	-2.87% slower
performance.compilation	286.05	274.5	1.04	4.04% faster
Mean result: 1.0044297086979403

---

Jenkins Console Log
Blue Ocean
Commit hash: 07501d859c49af3d678c3b786da70a7e77d8e698

---

Machine information

No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 20.04.3 LTS Release: 20.04 Codename: focal

CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
Address sizes: 46 bits physical, 48 bits virtual
CPU(s): 80
On-line CPU(s) list: 0-79
Thread(s) per core: 2
Core(s) per socket: 20
Socket(s): 2
NUMA node(s): 2
Vendor ID: GenuineIntel
CPU family: 6
Model: 85
Model name: Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz
Stepping: 4
CPU MHz: 3377.066
CPU max MHz: 3700.0000
CPU min MHz: 1000.0000
BogoMIPS: 4800.00
Virtualization: VT-x
L1d cache: 1.3 MiB
L1i cache: 1.3 MiB
L2 cache: 40 MiB
L3 cache: 55 MiB
NUMA node0 CPU(s): 0,2,4,6,8,10,12,14,16,18,20,22,24,26,28,30,32,34,36,38,40,42,44,46,48,50,52,54,56,58,60,62,64,66,68,70,72,74,76,78
NUMA node1 CPU(s): 1,3,5,7,9,11,13,15,17,19,21,23,25,27,29,31,33,35,37,39,41,43,45,47,49,51,53,55,57,59,61,63,65,67,69,71,73,75,77,79
Vulnerability Gather data sampling: Mitigation; Microcode
Vulnerability Itlb multihit: KVM: Mitigation: Split huge pages
Vulnerability L1tf: Mitigation; PTE Inversion; VMX conditional cache flushes, SMT vulnerable
Vulnerability Mds: Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Meltdown: Mitigation; PTI
Vulnerability Mmio stale data: Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed: Mitigation; IBRS
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; IBRS; IBPB conditional; STIBP conditional; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Mitigation; Clear CPU buffers; SMT vulnerable
Vulnerability Vmscape: Mitigation; IBPB before exit to userspace
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single pti intel_ppin ssbd mba ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req pku ospke md_clear flush_l1d arch_capabilities

G++:
g++ (Ubuntu 9.4.0-1ubuntu1~20.04) 9.4.0
Copyright (C) 2019 Free Software Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

Clang:
clang version 10.0.0-4ubuntu1
Target: x86_64-pc-linux-gnu
Thread model: posix
InstalledDir: /usr/bin

[avehtari]

Rebased onto develop now that #3402 has merged (force-pushed). Changes since this PR was opened:

• Rebased — the branch was based on pre-fix standalone gqs rng seed #3402 develop.
• Dropped the duplicated test helpers. The first version added its own bernoulli_fit_draws() and test_writer; fix standalone gqs rng seed #3402 now provides those (plus data_lines), so this PR adds only its test.
• Added genDraws_bernoulli_multi_chain_matches_single — chain i of a multi-chain run must reproduce a single-chain run at chain id 1 + i. It needs fix standalone gqs rng seed #3402 for the per-chain streams and this PR for the column alignment, so it could not pass on either branch alone and was held back until fix standalone gqs rng seed #3402 landed.
• Extended data_lines() to normalize the CSV separator: the Eigen writer path emits ", " where the std::vector path emits ",", and the new cross-path test is the first one to compare the two. The separator inconsistency itself remains out of scope here.

The fix in src/stan/services/util/gq_writer.hpp is unchanged.