ci: add scala-build prebuild job to share compilation across Scala CI jobs

[Yicong-Huang]

What changes were proposed in this PR?

The four sbt-based CI jobs — amber, amber-integration, platform, and platform-integration — each compile the shared common-module graph (DAO, WorkflowCore, WorkflowOperator, Auth, Config, …) from scratch. That is up to 14 runner legs redoing the same Scala build every run.

This adds a single scala-build prebuild job that compiles the whole Scala world once and hands the result to the downstream jobs as artifacts.

scala-build (ubuntu, one leg) runs one sbt invocation:

• scalafmtCheckAll + scalafixAll --check (cross-cutting lints, moved here from amber / amber-integration)
• all 7 <svc>/dist
• root Test/compile (compiles every module's test scope; TexeraProject aggregates all modules)

…then uploads two artifacts:

• scala-build-target — per-module target/ tree (classes, test-classes, Zinc analysis, generated JOOQ sources, dists). Restored by amber / amber-integration (Linux) / platform so their jacoco/test reuse compilation.
• scala-dists — just the dist zips, for platform-integration's smoke-boot.

Downstream jobs drop their own lint + compile and consume the artifacts.

Respecting the label gate: scala-build.if is the union of the four Scala run_* inputs, so a docs-only / frontend-only PR never runs it and every downstream Scala job stays skipped in lockstep. The required-checks aggregator depends on the whole build reusable-workflow result, so the new job is transparent to the merge gate (no .asf.yaml change). Artifact names are computed per build.yml invocation (from job_name_suffix / backport_target_branch) so the main build and the backport legs — which share one workflow run — don't collide on upload.

Known caveats / follow-ups:

• macOS amber-integration still self-compiles (Zinc analysis is not OS-portable). Because needs: is per-job, the macOS leg now waits for scala-build before starting; if the measured hit is large, split it into an independent job with no needs:.
• Zinc cross-job reuse relies on github.workspace being a stable path across GitHub-hosted ubuntu runners; if the analysis is ever rejected, sbt just recompiles — slower, never wrong.
• The 6-leg platform matrix each downloads scala-build-target; if download time dominates, slim the tar or split per service.

Primary goal is deduplicating compilation / cutting runner-minutes, not necessarily wall-clock — a serial prebuild + artifact transfer is inserted before the fan-out. Opened as draft to gather real numbers (wall-clock and total runner-minutes vs. a baseline) before committing to the approach.

Any related issues, documentation, discussions?

Closes #6318

How was this PR tested?

This is a CI-only change, validated by the workflow run on this PR itself:

• Confirm scala-build runs (the PR carries the ci label, which selects every stack) and the four downstream Scala jobs succeed off the restored artifacts.
• Confirm from the downstream sbt logs that jacoco/test do not trigger a full recompile (Zinc reuse working).
• Confirm the binary LICENSE/NOTICE checks in amber / platform and the platform-integration smoke-boot run off the restored dists.
• Locally: actionlint passes on .github/workflows/build.yml (no new findings); YAML parses clean.

build.yml is a reusable workflow with no unit-test harness, so correctness is verified end-to-end by the run rather than by added tests.

Was this PR authored or co-authored using generative AI tooling?

Generated-by: Claude Code (Claude Opus 4.8)

[github-actions]

Automated Reviewer Suggestions

Based on the git blame history of the changed files, we recommend the following reviewers:

• Contributors with relevant context: @mengw15
  You can notify them by mentioning @mengw15 in a comment.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 60.98%. Comparing base (0e92b16) to head (6fcf3e7).

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #6317      +/-   ##
============================================
- Coverage     61.02%   60.98%   -0.04%     
+ Complexity     3374     3373       -1     
============================================
  Files          1140     1140              
  Lines         44923    44923              
  Branches       4942     4942              
============================================
- Hits          27413    27396      -17     
- Misses        16001    16015      +14     
- Partials       1509     1512       +3     

Flag	Coverage Δ
access-control-service	70.00% <ø> (ø)
agent-service	44.59% <ø> (ø)
amber	66.39% <ø> (-0.01%)	⬇️
computing-unit-managing-service	0.00% <ø> (ø)
config-service	52.30% <ø> (ø)
file-service	63.97% <ø> (ø)
frontend	53.25% <ø> (-0.08%)	⬇️
notebook-migration-service	78.94% <ø> (ø)
pyamber	91.18% <ø> (ø)
workflow-compiling-service	55.14% <ø> (ø)

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

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[github-actions]

✅ No material benchmark regressions detected

🟢 4 better · 🔴 0 worse · ⚪ 11 noise (<±5%) · 0 without baseline

Compared against main 0e92b16 benchmarked on this same runner, so the delta is largely free of cross-runner hardware noise. The "7d avg" column still reflects the gh-pages dashboard. Treat <±5% as noise unless repeated.

Dashboard · Run

	config	throughput	MB/s	latency	max Δ latest / 7d
🟢	bs=10 sw=10 sl=64	398	0.243	24,395/34,262/34,262 us	🟢 -11.0% / 🔴 +131.6%
🟢	bs=100 sw=10 sl=64	804	0.49	125,193/139,886/139,886 us	🟢 -9.4% / 🔴 +35.6%
⚪	bs=1000 sw=10 sl=64	917	0.559	1,091,267/1,140,823/1,140,823 us	⚪ within ±5% / 🔴 +15.0%

Baseline details

Latest main 0e92b16 from same runner

config	metric	PR	latest main	7d avg	Δ latest	Δ 7d
bs=10 sw=10 sl=64	throughput	398 tuples/sec	405 tuples/sec	814.14 tuples/sec	-1.7%	-51.1%
bs=10 sw=10 sl=64	MB/s	0.243 MB/s	0.247 MB/s	0.497 MB/s	-1.6%	-51.1%
bs=10 sw=10 sl=64	p50	24,395 us	24,095 us	12,120 us	+1.2%	+101.3%
bs=10 sw=10 sl=64	p95	34,262 us	38,490 us	14,792 us	-11.0%	+131.6%
bs=10 sw=10 sl=64	p99	34,262 us	38,490 us	18,232 us	-11.0%	+87.9%
bs=100 sw=10 sl=64	throughput	804 tuples/sec	804 tuples/sec	1,050 tuples/sec	0.0%	-23.4%
bs=100 sw=10 sl=64	MB/s	0.49 MB/s	0.49 MB/s	0.641 MB/s	0.0%	-23.5%
bs=100 sw=10 sl=64	p50	125,193 us	119,830 us	96,227 us	+4.5%	+30.1%
bs=100 sw=10 sl=64	p95	139,886 us	154,436 us	103,174 us	-9.4%	+35.6%
bs=100 sw=10 sl=64	p99	139,886 us	154,436 us	111,033 us	-9.4%	+26.0%
bs=1000 sw=10 sl=64	throughput	917 tuples/sec	915 tuples/sec	1,072 tuples/sec	+0.2%	-14.5%
bs=1000 sw=10 sl=64	MB/s	0.559 MB/s	0.559 MB/s	0.655 MB/s	0.0%	-14.6%
bs=1000 sw=10 sl=64	p50	1,091,267 us	1,089,462 us	950,645 us	+0.2%	+14.8%
bs=1000 sw=10 sl=64	p95	1,140,823 us	1,154,337 us	991,936 us	-1.2%	+15.0%
bs=1000 sw=10 sl=64	p99	1,140,823 us	1,154,337 us	1,022,498 us	-1.2%	+11.6%

Raw CSV

config_idx,batch_size,schema_width,string_len,num_batches,total_ms,total_tuples,total_bytes,tuples_per_sec,mb_per_sec,lat_p50_us,lat_p95_us,lat_p99_us
0,10,10,64,20,503.02,200,128000,398,0.243,24395.19,34262.23,34262.23
1,100,10,64,20,2489.07,2000,1280000,804,0.490,125192.50,139885.58,139885.58
2,1000,10,64,20,21820.36,20000,12800000,917,0.559,1091267.15,1140822.68,1140822.68