CP-29639: Bump github.com/parquet-go/parquet-go from 0.30.1 to 0.32.0

[dependabot]

Bumps github.com/parquet-go/parquet-go from 0.30.1 to 0.32.0.

Release notes

Sourced from github.com/parquet-go/parquet-go's releases.

v0.32.0

Breaking changes

The format package now models thrift unions as sum types (parquet-go/parquet-go#558). If you use the format package directly:

• format.LogicalType and format.TimeUnit no longer expose one pointer field per union member (e.g. LogicalType.UTF8, TimeUnit.Millis); they now hold a single Value field typed as the format.LogicalTypeValue / format.TimeUnitValue interface.
• Some optional slice fields (e.g. level histograms in the column and offset indexes) changed from plain slices to thrift.Slice[T], which distinguishes unset from empty.

Code that only uses the top-level parquet package API is unaffected.

What's Changed

Bug fixes

• perf: extend WriteRowGroup fast paths (SortingWriter, bloom copy, buffers) + fix dedup/conversion bypass by @​achille-roussel in parquet-go/parquet-go#567

Other changes

• feat: verbatim column-chunk copy in Writer.WriteRowGroup by @​achille-roussel in parquet-go/parquet-go#535
• file: stop cloning and re-reading every page index chunk by @​achille-roussel in parquet-go/parquet-go#559
• feat(variant): columnar cursor reader for variant columns by @​ryanworl in parquet-go/parquet-go#552
• column: allocate the column tree from slabs by @​achille-roussel in parquet-go/parquet-go#560
• format: use thrift.Slice, thrift.Null, and model unions as sum types by @​achille-roussel in parquet-go/parquet-go#558
• perf(writer): enable L3 merge path for writers with bloom filters by @​T-J-L in parquet-go/parquet-go#562
• feat: Recurse the list tag through nested slices implicitly by @​starbaser in parquet-go/parquet-go#565
• column: cut column paths from a shared slab by @​achille-roussel in parquet-go/parquet-go#563
• perf(merge): replace heap with loser tree in mergedRowReader by @​achille-roussel in parquet-go/parquet-go#568
• fix: don't panic reading FIXED_LEN_BYTE_ARRAY decimal into []byte field by @​achille-roussel in parquet-go/parquet-go#570
• fix: MergeRowGroups produces unsorted output for descending sorting columns by @​achille-roussel in parquet-go/parquet-go#569
• test(merge): verify merged rows survive page buffer reuse by @​achille-roussel in parquet-go/parquet-go#571
• perf(merge): detect runs in the loser tree to skip repeated replays by @​achille-roussel in parquet-go/parquet-go#572
• perf(merge): bulk run emission with galloping and growable read buffers by @​achille-roussel in parquet-go/parquet-go#573
• feat(variant): streaming columnar writer for variant columns by @​ryanworl in parquet-go/parquet-go#553
• Fix index out of range panic in multiColumnChunk.OffsetIndex on merged row groups by @​SAY-5 in parquet-go/parquet-go#576
• perf(bounds): add AVX-512VL INT64 kernel for default-page window by @​perfloop-agent in parquet-go/parquet-go#579
• perf(reconstruct): reduce allocations for wide map reconstruction by @​perfloop-agent in parquet-go/parquet-go#581
• fix(writer): later rowgroups can overwrite earlier page counts by @​MichaHoffmann in parquet-go/parquet-go#577
• fix(writer): correctly size bloom filters on row group limit by @​T-J-L in parquet-go/parquet-go#578
• perf(format): zero-allocation footer decoding with format.FooterDecoder by @​ryanworl in parquet-go/parquet-go#556
• perf(merge): row-range overlap refinement by @​achille-roussel in parquet-go/parquet-go#574
• fix(value): format doubles with 64-bit precision in Value.String() by @​achille-roussel in parquet-go/parquet-go#583

New Contributors

• @​starbaser made their first contribution in parquet-go/parquet-go#565
• @​SAY-5 made their first contribution in parquet-go/parquet-go#576

Full Changelog: parquet-go/parquet-go@v0.31.0...v0.32.0

v0.31.0

What's Changed

... (truncated)

Commits

• 925bf65 fix(value): format doubles with 64-bit precision in Value.String() (#583)
• da3da86 perf(merge): row-range overlap refinement (#574)
• bd0bd83 perf(format): zero-allocation footer decoding with format.FooterDecoder (#556)
• e9317ce fix(writer): correctly size bloom filters on row group limit (#578)
• f28a62b fix(writer): later rowgroups can overwrite earlier page counts (#577)
• 3e52054 perf(reconstruct): reduce allocations for wide map reconstruction (#581)
• c215a4a perf(bounds): add AVX-512VL INT64 kernel for default-page window (#579)
• ef5d53a Fix index out of range panic in multiColumnChunk.OffsetIndex on merged row gr...
• fc63132 feat(variant): streaming columnar writer for variant columns (#553)
• 988b7b9 perf(merge): bulk run emission with galloping and growable read buffers (#573)
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

[greptile-apps]

PR author is in the excluded authors list.

[dependabot]

Looks like github.com/parquet-go/parquet-go is up-to-date now, so this is no longer needed.