[ancestral] error when writing gzipped VCF output

[jameshadfield]

Running augur ancestral with a gzipped output --output-vcf fileName.vcf.gz results in

Traceback (most recent call last):
  File "/Users/enigma/projects/nextstrain/augur/augur/__init__.py", line 66, in run
    return args.__command__.run(args)
  File "/Users/enigma/projects/nextstrain/augur/augur/ancestral.py", line 411, in run
    write_vcf(tree_dict, args.output_vcf)
  File "/Users/enigma/projects/treetime/treetime/vcf_utils.py", line 486, in write_vcf
    out_file.write( "##fileformat=VCFv4.2\n"+
  File "/Users/enigma/miniconda3/envs/augur/lib/python3.10/gzip.py", line 285, in write
    data = memoryview(data)
TypeError: memoryview: a bytes-like object is required, not 'str'

(Writing an uncompressed .vcf works fine.)