Update changelog

Lots of changes in this PR, but only a couple are user-facing. All other
changes can be considered performance improvements that do not change
functionality.

CHANGES.md

@@ -22,6 +22,8 @@
 * translate: Tree nodes are checked against the node-data JSON input to ensure sequences are present. [#1348][] (@jameshadfield)
 * export v2: Automatically minify large outputs. Use `--no-minify-json` to disable this default behavior. [#1352][] (@victorlin)
 * `augur.io.read_metadata`: A new optional `dtype` argument allows custom data types for all columns. Automatic type inference still happens by default, so this is not a breaking change. [#1252][] (@victorlin)
+* `augur.io.read_metadata`: A new optional `columns` argument allows specifying a subset of columns to load. The default behavior still loads all columns, so this is not a breaking change. [#1294][] (@victorlin)
+* filter: Add a new argument `--query-columns` allowing users to explicitly specify columns and data types for when automatic type inference behaves unexpectedly. [#1294][] (@victorlin)
 
 ### Bug Fixes
 
@@ -37,9 +39,11 @@
 * Deprecate `read_strains` from `augur.utils` and add it to the public API under `augur.io`. [#1353][] (@victorlin)
 * filter, frequencies, refine: Speed up reading of the metadata file. [#1252][] (@victorlin)
 * traits: Previously, columns with only numeric values were treated as numerical data. These are now treated as categorical data for discrete trait analysis. [#1252][] (@victorlin)
+* filter: The order of rows in `--output-metadata` and `--output-strains` now reflects the order in the original `--metadata`. [#1294][] (@victorlin)
 
 
 [#1252]: https://github.com/nextstrain/augur/pull/1252
+[#1294]: https://github.com/nextstrain/augur/pull/1294
 [#1344]: https://github.com/nextstrain/augur/pull/1344
 [#1348]: https://github.com/nextstrain/augur/pull/1348
 [#1351]: https://github.com/nextstrain/augur/pull/1351