Add basic subworkflow for unzipping and indexing fasta

[DLBPointon]

A pretty simple pattern i've seen in alot of places is unzipping fasta, cleaning and then index/sizes generation.

Thought I'd make it a subworkflow seeing as it's so common.

Linked to: sanger-tol/nf-core-modules#274

Further dev:

• Could potentially add something to clean headers too, alot of the tools we use don't like headers with any special characters or even spaces in a couple of cases.

EDIT: Just realised that it's set up for Sanger-tol data. I'll get that sorted.

[prototaxites]

Suggested change

	workflow GUNZIP_FASTA_INDEX {
	workflow FASTA_UNZIP_UPPER_GAWK {

Following the naming guidelines: https://nf-co.re/docs/specifications/components/subworkflows/naming-conventions#name-format-of-subworkflow-files

Anyone else have a better suggestion? I'm not used to the subworkflow naming conventions

[SPPearce]

There is a #subworkflows channel on the slack, good to ask there.
I think maybe something more general like FASTA_PREPARE

[DLBPointon]

Just spoke with Matthieu

Adding a header cleaner is likely a good idea, we already have a tool to do this in ASCC. Could also include some basic information about the fasta too (max_seq_length, total_length, n_seqs) in json format so that we don't pollute the meta.

[SPPearce]

Maybe we could generate the dict file as well, for the tools that require that.