Adds rules, config, and GitHub Action file to support running Nextclade
on all available sequences. Not yet tested.

Remove unnecessary configuration parameters from the Nextclade build
config and update the workflow to allow these parameters to be missing.
Since Snakemake evaluates the Python code in each rule's inputs,
outputs, and params, rules that we don't plan to run in the workflow can
produce key errors when their config parameters are not defined.

Simplifies the logic to get Nextclade datasets by following the same
pattern as the flu_frequencies workflow [1] where we grab the default
dataset for a given lineage and segment instead of specifying a
reference name. The "broad" and more recent references for H3N2 HA, for
example, are not too different from each other, but the Nextclade
annotations for the former are far more verbose than for the latter. We
also want the files produced by this workflow to plug directly into the
flu_frequencies workflow logic, so it is best to use the same approach
here.

[1] https://github.com/nextstrain/flu_frequencies/blob/6e4298fac3361f4a6751d85bcb963064dbb9eee1/Snakefile#L95

Adds NA to list of segments, since we want to know the subclade
annotations for NA as well as HA and use these data to estimate
frequencies.

Add temporary trigger for Nextclade workflow on PR event. This should
trigger the workflow when I push the update to the PR. If it works, I
should drop this commit again.

The workflow ran successfully, so removing this trigger.

Reduce threads requested for Nextclade runs from 16 to 12 so we can run 3 Nextclade jobs at once (one per lineage) on a 36-core instance of AWS Batch.