[yangzhang75]

What changes were proposed in this PR?

Everything so far has been machinery with nothing to drive it. This is the control: in the share
dialog, a published workflow gets a two-way switch for what the public sees.

• Two states, so the control is a switch. Follow latest is what Texera does today — the
  public sees the author's latest, updated on every save. Pinned holds the version they froze.
  Both choices stay on screen, so the author reads what the options are instead of inferring one
  from the label of the other.
• Everything under the switch describes the side it is on — a fact, not a warning: keeping an
  older version public is a legitimate choice. Green while the public has everything, amber once a
  pin is holding a version back.
• Only the pinned side has more to say. Behind, that is the one decision the state leaves
  open, so it gets a card: which version is out there, what it costs, and the act that ends it.
• The panel describes the saved copy, and the editor saves on a debounce, so it re-reads when
  a save lands. Deliberately it does not force a save first: the canvas is not always the workflow
  — it is empty while one loads, and stays empty if the collaborative model never arrives — so a
  save nobody asked for could write that emptiness over every operator the workflow had.

Behind a flag, off. gui.workflow-workspace.version-pinning-enabled defaults to false, so
merging this changes nothing anyone can see: the dialog is exactly what it is today and does not
even ask the server about a state it cannot show. The last PR of the series turns it on, so the
feature appears only once every surface that has to agree with a pin is in place.

Any related issues, documentation, discussions?

Closes #7942
Part of #7828. Design discussion: #7128. Stacked on #8576 and #8577; until those merge this
PR shows their commits too, and the review here is the last one.

How was this PR tested?

Frontend: the share-access spec covers the panel end to end — each of the three states and the
sentence it shows, the switch calling pin and unpin and doing nothing when the side already in
force is picked, "Update to current" asking for a re-pin, the date the card names, failures
surfacing as a notification, the panel staying hidden for a workflow that is not published and
for a user who cannot publish, and the re-read when a save lands. One case covers the flag being
off: nothing renders and the status endpoint is not called. The persist spec covers the new
pin/unpin/status calls and that a save is announced when it lands rather than when it is sent.
Full suite passes (6602), production (AOT) build passes, prettier clean.

Backend: GuiConfigSpec pins the flag's default to off (72 tests); the dashboard package passes
(572); scalafmt clean.

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

Generated-by: Claude Code (Opus 5)

🤖 Generated with Claude Code

[github-actions]

Automated Reviewer Suggestions

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

• Contributors with relevant context: @aglinxinyuan, @tanishqgandhi1908, @mengw15
  You can notify them by mentioning @aglinxinyuan, @tanishqgandhi1908, @mengw15 in a comment.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 97.54902% with 5 lines in your changes missing coverage. Please review.
✅ Project coverage is 92.34%. Comparing base (3726dd7) to head (0332da9).
⚠️ Report is 29 commits behind head on main.

Files with missing lines	Patch %	Lines
...shboard/user/workflow/WorkflowPublishService.scala	94.87%	1 Missing and 1 partial ⚠️
...ra/web/resource/dashboard/SearchQueryBuilder.scala	50.00%	1 Missing ⚠️
...esource/dashboard/WorkflowSearchQueryBuilder.scala	97.95%	1 Missing ⚠️
...ponent/user/share-access/share-access.component.ts	97.72%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #8578      +/-   ##
============================================
+ Coverage     92.33%   92.34%   +0.01%     
  Complexity     5045     5045              
============================================
  Files          1260     1260              
  Lines         54067    54240     +173     
  Branches       6755     6781      +26     
============================================
+ Hits          49921    50090     +169     
- Misses         2486     2489       +3     
- Partials       1660     1661       +1     

Flag	Coverage Δ		*Carryforward flag
access-control-service	77.38% <ø> (ø)
agent-service	99.16% <ø> (ø)		Carriedforward from 3726dd7
amber	87.37% <96.80%> (+0.05%)	⬆️
computing-unit-managing-service	60.48% <ø> (ø)
config-service	87.50% <100.00%> (+0.12%)	⬆️
file-service	82.00% <ø> (ø)
frontend	96.55% <98.71%> (+<0.01%)	⬆️
notebook-migration-service	83.73% <ø> (ø)
pyamber	98.57% <ø> (ø)		Carriedforward from 3726dd7
workflow-compiling-service	72.43% <ø> (ø)

*This pull request uses carry forward flags. Click here to find out more.

☔ 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]

⚠️ Benchmark changes need a look

🟢 4 better · 🔴 1 worse · ⚪ 10 noise (<±5%) · 0 without baseline

Compared against main 3726dd7 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	576	0.351	17,186/23,505/23,505 us	🟢 -20.0% / 🔴 +66.5%
🟢	bs=100 sw=10 sl=64	1,240	0.757	80,921/88,155/88,155 us	🟢 -12.7% / 🟢 -18.8%
⚪	bs=1000 sw=10 sl=64	1,395	0.851	711,619/806,852/806,852 us	⚪ within ±5% / 🟢 -12.3%

Baseline details

Latest main 3726dd7 from same runner

config	metric	PR	latest main	7d avg	Δ latest	Δ 7d
bs=10 sw=10 sl=64	throughput	576 tuples/sec	595 tuples/sec	968.33 tuples/sec	-3.2%	-40.5%
bs=10 sw=10 sl=64	MB/s	0.351 MB/s	0.363 MB/s	0.591 MB/s	-3.3%	-40.6%
bs=10 sw=10 sl=64	p50	17,186 us	15,944 us	10,646 us	+7.8%	+61.4%
bs=10 sw=10 sl=64	p95	23,505 us	29,376 us	14,120 us	-20.0%	+66.5%
bs=10 sw=10 sl=64	p99	23,505 us	29,376 us	16,498 us	-20.0%	+42.5%
bs=100 sw=10 sl=64	throughput	1,240 tuples/sec	1,250 tuples/sec	1,252 tuples/sec	-0.8%	-1.0%
bs=100 sw=10 sl=64	MB/s	0.757 MB/s	0.763 MB/s	0.764 MB/s	-0.8%	-0.9%
bs=100 sw=10 sl=64	p50	80,921 us	79,363 us	83,736 us	+2.0%	-3.4%
bs=100 sw=10 sl=64	p95	88,155 us	100,957 us	92,464 us	-12.7%	-4.7%
bs=100 sw=10 sl=64	p99	88,155 us	100,957 us	108,561 us	-12.7%	-18.8%
bs=1000 sw=10 sl=64	throughput	1,395 tuples/sec	1,381 tuples/sec	1,321 tuples/sec	+1.0%	+5.6%
bs=1000 sw=10 sl=64	MB/s	0.851 MB/s	0.843 MB/s	0.806 MB/s	+0.9%	+5.6%
bs=1000 sw=10 sl=64	p50	711,619 us	722,244 us	811,260 us	-1.5%	-12.3%
bs=1000 sw=10 sl=64	p95	806,852 us	775,268 us	847,636 us	+4.1%	-4.8%
bs=1000 sw=10 sl=64	p99	806,852 us	775,268 us	879,277 us	+4.1%	-8.2%

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
0,10,10,64,20,347.44,200,128000,576,0.351,17185.86,23505.16,23505.16
1,100,10,64,20,1612.35,2000,1280000,1240,0.757,80920.52,88154.98,88154.98
2,1000,10,64,20,14337.84,20000,12800000,1395,0.851,711619.35,806851.80,806851.80