Chore
* chore: fix upload of coverage artifacts ([`d12422f`](https://github.com/kalekundert/macromol_dataframe/commit/d12422ff0869cff5aed4b057d4f0ab0d48103fe7))
Feature
* feat: parse non-sequential residue &34;indices&34;
mmCIF files can specify residue indices that are not sequential. The
purpose of these indices is to allow canonical residue numbers to be
used for whole families of proteins that may differ in their exact
sequences. The non-sequential indices are specified by both the
`auth_seq_id` (author sequence id) and `pdbx_PDB_ins_code` (insertion
code) fields.
I made the decision that these fields are only useful together, so
there&39;s no point in providing them separately. ([`d7a38e5`](https://github.com/kalekundert/macromol_dataframe/commit/d7a38e5bbcdde96297c387d371a35ba83dd81dd3))
* feat: parse the `_entity_poly` table ([`41d0ebb`](https://github.com/kalekundert/macromol_dataframe/commit/41d0ebb17be0b57047d7e6f49be6f94c6244905b))
Fix
* fix: help old versions of polars infer column dtype ([`0f59dcc`](https://github.com/kalekundert/macromol_dataframe/commit/0f59dcc7cc5564218261be80a99740be6d5ce743))