feat(athena): Part 2 — fill the parse-path todo!() arms and the config alias

[aoelvp94]

Part of #16252. Related: #13822. Independent PR; the sequence is at the bottom.

Problem

With an athena profile parsing (#16274), dbt parse panics at the first todo!("Athena") it reaches, relation/factory.rs. Five of the nine todo!("Athena") arms are on the parse path; the other four are metadata, execution and dbt init, handled in later parts.

Solution

Fills the five parse-path arms. Each is the smallest thing that is correct for Athena, with the reason in a comment at the site:

• relation/factory.rs: Athena joins the generic RelationStatic arm with the other plain-SQL adapters. Same three lines as feat(athena): AthenaDbConfig profile schema #16274; whichever merges second drops them on rebase.
• adapter_impl.rs: the schema-listing column is schema_name, since Athena's information_schema is Trino's.
• column/column_builder.rs (two arms): build_athena, structurally identical to build_exasol, delegating every type decision to sql_types, which already carries ATHENA_KEYS. Deliberately not routed through build_postgres_like, whose Timestamp -> "datetime" rendering that path documents as a legacy quirk.
• relations/base.rs: microbatch event-time boundaries as TIMESTAMP '...' literals. The generic arm emits a bare varchar, which Trino rejects (Cannot apply operator: timestamp >= varchar); Trino's literal takes no T and no zone offset.
• seed_io.rs: column-name inference is Lowercase, because the Glue catalog stores column names lowercase whatever the DDL says.

adapter.convert_type also returns dbt-athena's names for Athena: convert_text_type gives string and convert_number_type gives integer for whole numbers, with no width distinction. The macros compare against those names, so returning the DDL spellings varchar and bigint made an insert_overwrite model partitioned on a text or bigint column fail on its second run in get_partition_batches. Mapped at the convert_type boundary rather than in format_arrow_type_as_sql, which renders unit-test fixtures and seed DDL and must keep the Trino spellings and the integer widths.

Plus config_aliases: dbt-athena 1.11.0 declares _ALIASES = {"catalog": "database"} (connections_legacy.py), which canonicalises node config keys. Athena leaves the "no aliases" group and gets its own arm.

The two remaining adapter todo!("Athena") arms (MetadataAdapter construction, get_relation) are annotated, not filled; Part 5 fills them.

Verification

• cargo fmt --check, cargo clippy on dbt-adapter, dbt-adapter-core, dbt-df-providers, dbt-schemas clean on the pinned toolchain.
• With Parts 1, 3 and 4 applied on top, dbt parse on a 750-model project completes with 0 errors in 1.8 s; dbt ls exits 0. Without Part 4 it stops at the missing dbt-athena package, as feat(athena): AthenaDbConfig profile schema #16274 describes.

Feedback wanted: build_athena mirrors build_exasol rather than build_postgres_like; confirm that is the intended direction for Trino-family adapters.

Sequence (independent PRs, each compiles alone against main)

• Part 0: feat(athena): Part 0 — map Backend::Athena to its driver entrypoint and send the driver's output-location option #16365 — ADBC driver wiring
• Part 1: feat(athena): AthenaDbConfig profile schema #16274 — AthenaDbConfig profile schema (@fpiped)
• Part 2: this PR — dbt-adapter core wiring
• Part 3: feat(athena): Part 3 — accept partitioned_by, table_properties and force_batch as model configs #16368 — model config keys
• Part 4: feat(athena): Part 4 — vendor the dbt-athena 1.11.0 macro package #16369 — dbt-athena macro package
• Part 5: feat(athena): Part 5 — AthenaMetadataAdapter and athena_get_relation #16374 — AthenaMetadataAdapter and get_relation
• Parts 6+7: branch ready, not yet opened — Glue / Lake Formation housekeeping shims and the table / incremental helpers (athena/parts-6-7-execution); see also feat(athena): Part 6 — AthenaAdapter methods behind the dbt-athena macros #16376
• Part 8: feat(athena): Part 8 — run dbt unit tests on Athena #16370 — dbt unit tests on Athena
• Part 9: feat(athena): Part 9 — dbt init profile wizard for Athena #16375 — dbt init profile wizard
• Full branch for context: https://github.com/aoelvp94/dbt/tree/spike/athena-profile-parse

Checklist

• I have read the contributing guide and understand what's expected of me.
• I have run this code in development, and it appears to resolve the stated issue.
• This PR includes tests, or tests are not required or relevant for this PR. (unit tests for the column builder and the timestamp literal are a pending follow-up on this branch)
• This PR has no interface changes.

[fpiped]

Tested merged with the rest of the stack against a real workgroup: debug, compile --write-catalog, docs generate work, and with Part 6 the materializations do too.

One divergence that bites: convert_type returns varchar and bigint for Athena where AthenaAdapter.convert_text_type / convert_number_type return string and integer. delete_overlapping_partitions and get_partition_batches only handle integer / string / date / timestamp, so an insert_overwrite model partitioned by a text or bigint column fails with Need to add support for column type varchar on its second run. The seed macros already map string to varchar themselves, so returning dbt-athena's names is safe there.

Landing note: #16374 touches the same Athena arms in adapter_impl.rs and get_relation.rs.