Enable pangolin to read from stdin

[matt-sd-watson]

This aims to resolve #329.

pangolin can now read in stdin from a bash pipe with "-", as requested in #329. If no query fasta or stdin is specified, such as :
pangolin --outfile lineage_report.csv
pangolin will output:
Error: input query fasta could not be detected from a filepath or through stdin. Please enter your fasta sequence file and refer to pangolin usage at: https://cov-lineages.org/pangolin.html for detailed instructions.

If "-" is used to designate stdin, but there is no stdin for pangolin to use, i.e. if nothing is piped into the pangolin command:
pangolin --outfile lineage_report.csv -

it will output:
Error: cannot find query (input) fasta file using stdin. Please enter your fasta sequence file and refer to pangolin usage at: https://cov-lineages.org/pangolin.html for detailed instructions.

The methods for dealing with file paths that do not exist remain the same as before.

If the user attempts to pass a compressed stdin, such as:
cat test.fa | gzip | pangolin --outfile lineage_report.csv -

pangolin will report:

Error: error when reading query fasta. It is possible that compressed stdin was passed.

A important note that is this error message above will be thrown if any of the inputs (compressed or not) cannot ultimately be read as a FASTA file by SeqIO. This may want to be changed in the future but I tried my best to allow pangolin to differentiate between stdin and filepaths and the compression status of either input.

[matt-sd-watson]

I made some changes based on the errors given by the initial pytest, and have tested extensively with various compressed and uncompressed inputs to verify the proper behaviour and desired error handling. Everything seems to work as intended on my end but let me know if there see to be any glaring problems!

[aineniamh]

This looks really great! Just started the tests running, but when they complete I'll merge in! Thanks for this @matt-sd-watson 🎉