phylogenetic: trim down workflow logs

[joverlee521]

I was curious why the last phylogenetic workflow failed and was digging into the logs and saw the run-build job's logs were truncated.

We might want to revisit the logging within the phylogenetic workflow and see if we can clean up any excess logs.
Two main candidates from my brief digging:

1. insertion-metadata.py currently prints out index_with_insertion within a loop, resulting in hundreds of lines in the logs that just building up the same list. This looks like a debugging print statement during development that can be removed entirely OR moved out the for-loop to only print the final list.

hmpv/phylogenetic/scripts/insertion-metadata.py

Lines 30 to 35 in f62053b

	index_with_insertion = []
	for index, row in alignedinsertionsdf.iterrows():
	has_insertion = any(len(str(row[col])) > 30 for col in column_with_insertion)
	if has_insertion:
	index_with_insertion.append(index)
	print(index_with_insertion)

2. augur align prints out insertions that flood the logs with the long insertion sequences. We might want to improve this in Augur (proposed in align: trim down print statements for insertions augur#1737), but for now we can redirect the stdout from the command into a log file.