MultiQC report not displaying all all samples for bakta/prokka

Description of the bug

The MultiQC report of a pipeline run currently shows only a single (randomly chosen?) sample from all samples in the samplesheet, e.g. in a pipeline run with sample_1 and sample_2, only one of them is reported:

Partial solution

Other pipelines also reported and solved this, thus part of the fix can be found in: https://github.com/nf-core/bacass/pull/134.
This will work for AMP/ARG/CAZyme subworkflows, but not for BGC subworkflow: Before executing annotations when BGC subworkflow is activated, the input sequences are split into two categories: "all" and "long" sequences, the latter containing only seqs above the length threshold (param --bgc_mincontiglength) for BGC screening. Since the sequences of both categories are annotated, we obtain two annotation log files per sample. Both those files have the same name (e.g. sample_2_prokka.txt), so MultiQC cannot discriminate them and (I think?) overwrites statistics from the first file with those from the second file.

Desired behaviour

The stats should be reported for all samples and in the case of BGC subworkflow for all samples and their two categories, e.g.:

• sample_1_prokka_all
• sample_1_prokka_long
• sample_2_prokka_all
• sample_2_prokka_long
• ...

(The _prokka is not necessary actually, but currently part of the log file name.)

Command used and terminal output

nextflow run nf-core/funcscan -r dev -profile docker,test_minimal --annotation_tool prokka --run_bgc_screening --bgc_skip_gecco false --outdir results

Relevant files

No response

System information

No response

Start by running the reported Nextflow command with the test_minimal profile and inspect how annotation log files from the AMP/ARG/CAZyme and BGC subworkflows reach MultiQC. Compare the existing approach with nf-core/bacass PR 134, then verify that all samples and both BGC categories appear separately in the report.