bases2fastq doesn't use generate_fastq_meta so fcid is null

[edmundmiller]

Description of the bug

example output:

tree null/
null/
├── 20220601_PLT-03_BBS-0174_1.fastp.fastq.gz
├── 20220601_PLT-03_BBS-0174_1_fastqc.html
├── 20220601_PLT-03_BBS-0174_1_fastqc.zip
├── 20220601_PLT-03_BBS-0174_2.fastp.fastq.gz
├── 20220601_PLT-03_BBS-0174_2_fastqc.html
├── 20220601_PLT-03_BBS-0174_2_fastqc.zip
├── 20220601_PLT-03_BBS-0174.fastp.html
├── 20220601_PLT-03_BBS-0174.fastp.json
├── 20220601_PLT-03_BBS-0174.fastp.log
├── DefaultSample_R1.fastq.gz.md5
├── DefaultSample_R2.fastq.gz.md5
├── Unassigned_R1.fastq.gz.md5
└── Unassigned_R2.fastq.gz.md5```

### Command used and terminal output

_No response_

### Relevant files

_No response_

### System information

_No response_

[matthdsm]

Shouldn't b2fq have a custom generate_fastq_meta? I think that might be valuable downstream, especially if we want to convert everything to subworkflows.

I haven't tested b2fq. Am I correct that there's no real meta / sample for the output?

[matthdsm]

fixed in #63