Standardized multiple inputs

## Description of proposed changes Use latest guidance for handling inputs based on https://github.com/nextstrain/pathogen-repo-guide/pull/91. - use shared `remote_inputs.smk`, which bumps minimum Snakemake to 8.0.0 - use `augur merge` for sequences - use `augur read-file` or `augur merge` depending on number of inputs (see f5c9949a810004b34b605f947622c7f26ec0910e) The config parameters remain unchanged so this should not affect outside users. Downstream rules are now expected to use "results/metadata.tsv" and "results/sequences_{segment}.fasta" as inputs instead of the input_* functions. ## Related issue(s) https://github.com/nextstrain/public/issues/25 ## Checklist <!-- Make sure checks are successful at the bottom of the PR. If applicable, add: - any changes to existing tests - any additional manual testing to confirm changes Please add a note if you need help with adding tests. --> - [x] Checks pass Post-merge: - [ ] https://github.com/nextstrain/augur/pull/1924 - [ ] https://github.com/nextstrain/docker-base/pull/271 - [ ] https://github.com/nextstrain/conda-base/pull/133 <!-- 🙌 Thank you for contributing to Nextstrain! ✨ -->