Use ALL_MODELS rather than DEMO_MODELS

[mhauru]

We define a number of models in TestUtils. Most of them are collected in the constant DEMO_MODELS, namely all the ones representing the one model with an InverseGamma and all that. However, there are a few other models as well. This PR collects all of them in a new constant ALL_MODELS, and starts using that in the test suite wherever the test isn't specific to the particular statistical model that DEMO_MODEL implements. It also fixes various issues that this expansion of the test suite brought up.

That this wasn't done before is the reason that in #1150 the issue with ranges wasn't caught by the test suite.

Not yet ready for review, need to understand remaining test failures.

[github-actions]

Benchmark Report

• this PR's head: d46bbd0642373929e3b32fad1f19600c12db5695
• base branch: 6266f644ce8caaa3b98fc65c1eb960f4f77243b1

Computer Information

Julia Version 1.11.7
Commit f2b3dbda30a (2025-09-08 12:10 UTC)
Build Info:
  Official https://julialang.org/ release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
  CPU: 4 × AMD EPYC 7763 64-Core Processor
  WORD_SIZE: 64
  LLVM: libLLVM-16.0.6 (ORCJIT, znver3)
Threads: 1 default, 0 interactive, 1 GC (on 4 virtual cores)

Benchmark Results

┌───────────────────────┬───────┬─────────────┬───────────────────┬────────┬─────────────────────────────────┬───────────────────────────┬─────────────────────────────────┐
│                       │       │             │                   │        │        t(eval) / t(ref)         │     t(grad) / t(eval)     │        t(grad) / t(ref)         │
│                       │       │             │                   │        │ ──────────┬───────────┬──────── │ ──────┬─────────┬──────── │ ──────────┬───────────┬──────── │
│                 Model │   Dim │  AD Backend │           VarInfo │ Linked │      base │   this PR │ speedup │  base │ this PR │ speedup │      base │   this PR │ speedup │
├───────────────────────┼───────┼─────────────┼───────────────────┼────────┼───────────┼───────────┼─────────┼───────┼─────────┼─────────┼───────────┼───────────┼─────────┤
│               Dynamic │    10 │    mooncake │             typed │   true │    367.90 │    373.17 │    0.99 │ 10.69 │   10.53 │    1.01 │   3931.66 │   3931.03 │    1.00 │
│                   LDA │    12 │ reversediff │             typed │   true │   2605.12 │   2662.06 │    0.98 │  5.09 │    5.16 │    0.99 │  13257.24 │  13736.49 │    0.97 │
│   Loop univariate 10k │ 10000 │    mooncake │             typed │   true │ 106259.67 │ 104295.49 │    1.02 │  4.13 │    4.24 │    0.97 │ 439060.60 │ 442337.88 │    0.99 │
├───────────────────────┼───────┼─────────────┼───────────────────┼────────┼───────────┼───────────┼─────────┼───────┼─────────┼─────────┼───────────┼───────────┼─────────┤
│    Loop univariate 1k │  1000 │    mooncake │             typed │   true │   7979.76 │   8008.68 │    1.00 │  4.71 │    4.78 │    0.99 │  37603.23 │  38302.78 │    0.98 │
│      Multivariate 10k │ 10000 │    mooncake │             typed │   true │  33284.46 │  75318.18 │    0.44 │ 10.06 │    6.14 │    1.64 │ 334891.43 │ 462645.91 │    0.72 │
│       Multivariate 1k │  1000 │    mooncake │             typed │   true │   3645.94 │   3593.73 │    1.01 │ 12.76 │    9.29 │    1.37 │  46516.54 │  33375.74 │    1.39 │
├───────────────────────┼───────┼─────────────┼───────────────────┼────────┼───────────┼───────────┼─────────┼───────┼─────────┼─────────┼───────────┼───────────┼─────────┤
│ Simple assume observe │     1 │ forwarddiff │             typed │  false │      2.75 │      2.69 │    1.02 │  3.94 │    4.01 │    0.98 │     10.83 │     10.78 │    1.00 │
│           Smorgasbord │   201 │ forwarddiff │             typed │  false │   1216.41 │   1222.03 │    1.00 │ 63.26 │   64.23 │    0.98 │  76947.07 │  78491.54 │    0.98 │
│           Smorgasbord │   201 │ forwarddiff │       simple_dict │   true │       err │       err │     err │   err │     err │     err │       err │       err │     err │
├───────────────────────┼───────┼─────────────┼───────────────────┼────────┼───────────┼───────────┼─────────┼───────┼─────────┼─────────┼───────────┼───────────┼─────────┤
│           Smorgasbord │   201 │ forwarddiff │ simple_namedtuple │   true │       err │       err │     err │   err │     err │     err │       err │       err │     err │
│           Smorgasbord │   201 │      enzyme │             typed │   true │   1673.06 │   1690.43 │    0.99 │  6.76 │    5.40 │    1.25 │  11313.27 │   9136.45 │    1.24 │
│           Smorgasbord │   201 │    mooncake │             typed │   true │   1677.76 │   1685.75 │    1.00 │  5.28 │    5.40 │    0.98 │   8852.73 │   9110.41 │    0.97 │
├───────────────────────┼───────┼─────────────┼───────────────────┼────────┼───────────┼───────────┼─────────┼───────┼─────────┼─────────┼───────────┼───────────┼─────────┤
│           Smorgasbord │   201 │ reversediff │             typed │   true │   1697.35 │   1743.22 │    0.97 │ 86.95 │   89.83 │    0.97 │ 147586.82 │ 156586.08 │    0.94 │
│           Smorgasbord │   201 │ forwarddiff │      typed_vector │   true │   1676.98 │   1701.53 │    0.99 │ 59.95 │   58.92 │    1.02 │ 100530.75 │ 100257.95 │    1.00 │
│           Smorgasbord │   201 │ forwarddiff │           untyped │   true │   1687.95 │   1688.86 │    1.00 │ 64.85 │  125.87 │    0.52 │ 109468.97 │ 212569.83 │    0.51 │
├───────────────────────┼───────┼─────────────┼───────────────────┼────────┼───────────┼───────────┼─────────┼───────┼─────────┼─────────┼───────────┼───────────┼─────────┤
│           Smorgasbord │   201 │ forwarddiff │    untyped_vector │   true │   1672.63 │   1705.46 │    0.98 │ 60.95 │   58.53 │    1.04 │ 101941.68 │  99822.96 │    1.02 │
│              Submodel │     1 │    mooncake │             typed │   true │      7.53 │      7.19 │    1.05 │  4.87 │    5.14 │    0.95 │     36.67 │     36.95 │    0.99 │
└───────────────────────┴───────┴─────────────┴───────────────────┴────────┴───────────┴───────────┴─────────┴───────┴─────────┴─────────┴───────────┴───────────┴─────────┘

[github-actions]

DynamicPPL.jl documentation for PR #1173 is available at:
https://TuringLang.github.io/DynamicPPL.jl/previews/PR1173/

[codecov]

Codecov Report

❌ Patch coverage is 58.33333% with 10 lines in your changes missing coverage. Please review.
✅ Project coverage is 78.90%. Comparing base (6266f64) to head (d46bbd0).

Files with missing lines	Patch %	Lines
src/threadsafe.jl	0.00%	8 Missing ⚠️
src/test_utils/model_interface.jl	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1173      +/-   ##
==========================================
- Coverage   78.95%   78.90%   -0.06%     
==========================================
  Files          41       41              
  Lines        3896     3910      +14     
==========================================
+ Hits         3076     3085       +9     
- Misses        820      825       +5     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[penelopeysm]

Otherwise looks great!