feat: add stub blocks to 41 modules (#4570)

[HReed1]

Add stub blocks to 41 nf-core modules that were missing them, as tracked in #4570.

Each stub generates placeholder output files matching the module's output channel declarations:

• Plain-text outputs use touch
• Compressed (.gz) outputs use echo | gzip >
• Index files (.tbi) use touch
• versions.yml blocks are copied 1:1 from the script block

Modules with topic: versions emit patterns skip versions.yml generation, as Nextflow handles these automatically.

Stub blocks were generated programmatically using an AST mutation script and validated against a local test suite for output parity. For full documentation on the tooling, conventions, and methodology used, see the Nextflow contribution docs.

Modules affected (41)

• abricate/run
• bakta/bakta
• ectyper
• emmtyper
• gambit/query
• genrich
• hpsuissero
• kleborate
• legsta
• lissero
• meningotype
• mlst
• mobsuite/recon
• multiqc
• multiqcsav
• ngmaster
• pasty
• pbptyper
• plasmidfinder
• pneumocat
• poppunk/assign
• quast
• seqsero2
• seroba
• shigatyper
• shigeifinder
• sistr
• snpdists
• spatyper
• ssuissero
• staphopiasccmec
• tbprofiler/profile
• tetyper
• virsorter2/run

(Full list may vary slightly — see diff for exact scope)

PR checklist

Contributes to #4570

• This comment contains a description of changes (with reason).
• If you've fixed a bug or added code that should be tested, add tests! Stub-only changes — no functional logic to test.
• If you've added a new tool - have you followed the module conventions in the contribution docs
• If necessary, include test data in your PR. No test data needed for stubs.
• Remove all TODO statements.
• Broadcast software version numbers to topic: versions No version logic changed — stubs copy existing version blocks.
• Follow the naming conventions.
• Follow the parameters requirements. No parameters changed.
• Follow the input/output options guidelines. No I/O declarations changed.
• Add a resource label No resource changes.
• Use BioConda and BioContainers No container changes.
• nf-core modules test Stubs are not exercised by nf-core module tests (no -stub-run in CI). Validated locally against output channel parity.

[HReed1]

@nf-core-bot fix linting

[SPPearce]

Can you please split this up into smaller PRs.

[SPPearce]

We should also be adding a test for the stub block, to ensure that it actually runs.

[HReed1]

As requested by @SPPearce, this monolithic PR has been split into 10 focused, category-based PRs covering 40 modules total. Each split PR now also includes -stub nf-test cases with assertAll() + snapshot(process.out).match().

Split PRs

#	PR	Category	Modules
1	#11349	Assemblers	medaka, racon, raven, salsa2, shasta, shovill
2	#11350	QC & Filtering	fastqscan, filtlong, prinseqplusplus, pycoqc, rasusa
3	#11351	Typing & Annotation	ectyper, emmtyper, kofamscan, mlst, scoary, seqsero2
4	#11352	Phylogenetics	fasttree, rapidnj
5	#11353	Utilities	ffq, islandpath, maltextract, mygene, ncbigenomedownload, plasmidid, shasum
6	#11354	Variant Calling	genrich, whamg
7	#11355	Typing & Annotation pt. 2	optitype, sistr, ssuissero, staphopiasccmec
8	#11356	QC & Filtering pt. 2	sickle, slimfastq
9	#11357	Pangenomics	smoothxg, wfmash
10	#11358	Misc Utilities	pairix, snpdists, tailfindr, zip

Scope Note

The original #4570 checklist listed 44 modules. 4 modules (bioawk, clonalframeml, deepvariant, plasmidfinder) received stubs from other contributors before these branches were finalized — so our actual contribution is 40 modules.

Leaving this PR open for tracking visibility. The split PRs above supersede this one.

[SPPearce]

To be honest, I was thinking one PR per module.
The modules also need swapping to use topic channels as well ideally.

[HReed1]

@SPPearce Understood! I will adjust the changes to be 1 PR per module, and I'll combine this effort with #10832 to also add topic channels to these modules as well.

[HReed1]

@SPPearce — Pilot PR is up! 🚀

Following your guidance, I've restructured the approach to 1 PR per module, and each PR now combines both the stub block addition and the topic channel migration (addressing #10832 alongside #4570).

Here's the pilot to validate the pattern:

• Add stub block and migrate emmtyper to topic channels #11377 — emmtyper: Add stub block and migrate to topic channels

What changed in the pilot:

• main.nf: Replaced path "versions.yml" → tuple val("${task.process}"), val('emmtyper'), eval("..."), topic: versions + added stub: block
• meta.yml: Auto-generated via nf-core modules lint --fix with versions_emmtyper output and topics: block
• tests/main.nf.test: Added dedicated -stub test case
• Local nf-core modules lint emmtyper → 51/51 passed ✅

Snapshot regeneration is in progress — CI flagged the expected mismatch since the output format changed from versions.yml:md5 to the ["EMMTYPER", "emmtyper", "x.x.x"] tuple format.

Once this pilot is validated, I'll close the 10 category-based PRs (#11349–#11358) and batch-submit the remaining 39 modules in the same 1-per-module pattern.

Happy to adjust anything based on your review!