Migrate narfmap/hashtable to topic channels

[xHaMMaDy]

Description

• replace the legacy versions.yml output with a versions_narfmap tuple broadcast to topic: versions
• use dragen-os --version 2>&1 to report the NARFMAP version
• update the module metadata, nf-tests, and generated snapshots for the structured version output

Generated by Codex

PR checklist

Closes #12319

• This comment contains a description of changes (with reason).
• Tests cover the changed version output.
• Remove all TODO statements.
• Broadcast software version numbers to topic: versions.
• Follow the naming conventions.
• Follow the parameters requirements.
• Follow the input/output options guidelines.
• Add a resource label.
• Use BioConda and BioContainers if possible to fulfil software requirements.

Tests

• nf-core modules lint narfmap/hashtable (59 passed)
• Docker nf-test (2 passed)
• Singularity nf-test (2 passed)
• Micromamba nf-test using the Conda environment (2 passed)

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[LouisLeNezet]

Hi @xHaMMaDy,
Would you have some time to finish this PR ?

[LouisLeNezet]

This line should be as a separate assertion outside the snapshot.

[LouisLeNezet]

I meant

            assertAll (
                { assert process.success },
                { assert file(process.out.log[0][1]).readLines().findAll { it.startsWith("decompHash") } },
                { assert snapshot(
                    process.out.bam.collect { meta, bamFile -> [meta, file(bamFile).name, bam(bamFile).getReadsMD5()] },
                    process.out.findAll { key, val -> key.startsWith("versions") },
                ).match() }
            )

[LouisLeNezet]

You just need now to update the meta and snapshot

[LouisLeNezet]

I meant

            assertAll (
                { assert process.success },
                { assert file(process.out.log[0][1]).readLines().findAll { it.startsWith("decompHash") } },
                { assert snapshot(
                    process.out.bam.collect { meta, bamFile -> [meta, file(bamFile).name, bam(bamFile).getReadsMD5()] },
                    process.out.findAll { key, val -> key.startsWith("versions") },
                ).match() }
            )

[xHaMMaDy]

Pushed an update for the align test snapshots.

meta.yml already mirrors the simplified Samtools eval from main.nf.

I also tested the exact getReadsMD5() form locally for the segfault regression fixture, but it fails inside HTSJDK because the emitted BAM contains an invalid sequence name:

htsjdk.samtools.SAMException: Sequence name ... doesn't match regex

So for this specific malformed-BAM regression test, I moved the decompHash log check outside the snapshot, kept the BAM existence/non-empty checks outside the snapshot, and snapshot only the BAM meta/name plus versions. The normal BAM tests still use getReadsMD5().

Local checks:

• narfmap/align/tests/segfault.nf.test passes with Docker and Nextflow 25.10.2
• narfmap/align/tests/main.nf.test --update-snapshot passed and updated the normal BAM snapshots
• one full targeted local run hit a transient raw.githubusercontent.com staging timeout, and the isolated retry of that same shard passed

Comment added by Codex

[xHaMMaDy]

CI is green after the latest update.

I rebased the branch onto current master and kept the PR diff scoped to the NARFMAP modules. The align snapshots now avoid empty BAM MD5 values, and the hashtable snapshot uses stable output names with channel-aware sanitization.

Comment added by Codex