3.x: Remove Javadoc not failing on error

[benjamintboyle]

The 'javadoc' task should fail when there is an error. The
'javadocCleanup' task failed for PR 7239 due to publish plugin issue 242,
while it should have been the 'javadoc' task that failed the build.
Also, publishing when javadocs fail is not desired.

[codecov]

Codecov Report

Merging #7258 (91a17d6) into 3.x (06294de) will decrease coverage by 0.03%.
The diff coverage is n/a.

@@             Coverage Diff              @@
##                3.x    #7258      +/-   ##
============================================
- Coverage     99.57%   99.53%   -0.04%     
  Complexity     6758     6758              
============================================
  Files           747      747              
  Lines         47391    47391              
  Branches       6382     6382              
============================================
- Hits          47190    47172      -18     
- Misses           91       97       +6     
- Partials        110      122      +12     

Impacted Files	Coverage Δ	Complexity Δ
...l/operators/observable/ObservableFlatMapMaybe.java	92.25% <0.00%> (-4.93%)	2.00% <0.00%> (ø%)
...xjava3/internal/observers/FutureMultiObserver.java	96.61% <0.00%> (-3.39%)	27.00% <0.00%> (-1.00%)
.../operators/observable/ObservableFlatMapSingle.java	94.44% <0.00%> (-2.39%)	2.00% <0.00%> (ø%)
...va3/internal/operators/parallel/ParallelRunOn.java	98.53% <0.00%> (-1.47%)	8.00% <0.00%> (ø%)
...rnal/operators/flowable/FlowableSequenceEqual.java	98.78% <0.00%> (-1.22%)	2.00% <0.00%> (ø%)
...ternal/operators/observable/ObservableFlatMap.java	97.51% <0.00%> (-1.07%)	3.00% <0.00%> (ø%)
...3/internal/operators/observable/ObservableZip.java	99.31% <0.00%> (-0.69%)	6.00% <0.00%> (ø%)
.../operators/observable/ObservableCombineLatest.java	99.38% <0.00%> (-0.62%)	6.00% <0.00%> (ø%)
...operators/flowable/FlowableConcatMapScheduler.java	99.20% <0.00%> (-0.40%)	4.00% <0.00%> (ø%)
...a/io/reactivex/rxjava3/subjects/ReplaySubject.java	99.37% <0.00%> (-0.21%)	49.00% <0.00%> (ø%)
... and 3 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 06294de...91a17d6. Read the comment docs.

[akarnokd]

If I understand correctly, #7239 would override the JDK that compiles and tests RxJava to be 8 always. This is not desired because the purpose of the JDK 11 build is to verify the fix of #7173, which needs an actual JDK 11 runtime. We don't care about the Javadoc generated for the JDK 11 build, only the build of JDK 8.

[benjamintboyle]

Thanks for giving that history. I'll have to think through it now that I understand it.

I think that this pull request is valuable and independent of that though. Correct?

[benjamintboyle]

Were you implying that the Javadoc task for JDK 11 Workflow the was added in the previous pull request should be removed? That would make sense

[akarnokd]

I think that this pull request is valuable

I think it was disabled because pre-Java 8 javadoc generation sometimes failed even though it managed to do its job. Let's have this PR.

Were you implying that the Javadoc task for JDK 11 Workflow the was added in the previous pull request should be removed?

Let it be for now.