perf(parquet): reuse DataPageV2 assembly buffer for eager writes

[fallintoplace]

Rationale for this change

buildDataPageV2 creates a new bytes.Buffer for every page when joining levels with encoded values. Pages without active dictionary encoding are written synchronously, so their assembly storage does not need to survive the call. Reallocating it for every page adds page-sized allocation churn.

What changes are included in this PR?

• Reuse the column writer scratch buffer for eagerly written DataPageV2 pages.
• Keep an owned buffer for pages retained while dictionary encoding is active.
• Add a benchmark that writes 64 eager 64 KiB pages with uncompressed and Snappy codecs.

Medians from 6 runs on an Apple M1 Pro were:

codec	metric	main	this PR	change
uncompressed	time/op	1.08 ms	650 us	-39.6%
uncompressed	B/op	4,547,602	383,763	-91.6%
uncompressed	allocs/op	461	321	-30.4%
Snappy	time/op	9.05 ms	8.88 ms	-1.9%
Snappy	B/op	3,027,460	489,403	-83.8%
Snappy	allocs/op	459	325	-29.2%

Are these changes tested?

• go test ./parquet/... -count=1
• go test ./parquet/file -run "^$" -bench "^BenchmarkWriteDataPageV2Eager$" -benchmem -count=6

The Parquet suite includes multi-page DataPageV2 dictionary coverage for the retained-page path.

Are there any user-facing changes?

No. The Parquet output and public API are unchanged.