Add Comet 1.0.0 release announcement blog post [WIP - targeting release on 7/31]

[andygrove]

This PR adds a blog post announcing the Apache DataFusion Comet 1.0.0 release, scheduled for 2026-07-31.

The post follows the same format as previous Comet release announcements. It frames 1.0.0 as a maturity/GA milestone and summarizes the notable changes merged since the 0.17.0 tag, including:

• Native scalar expression performance optimizations across JSON, regex/URL, cast, decimal, date/time, and string/array kernels, plus Parquet metadata caching
• New expression and aggregate support (approx/exact percentile and median, approx_count_distinct, collect_list/array_agg, base64, levenshtein, randstr, interval types incl. CalendarIntervalType and multiply_dt_interval, grouping/grouping_id, array_prepend, split_part, MapType support)
• Experimental accelerated PyArrow UDFs
• Iceberg 1.11 support, Iceberg V3 native table decryption, and delete-file / exchange-reuse / native-serde correctness fixes
• gzip native Parquet writes and Azure authentication
• Execution and fallback improvements (mixed partial/final aggregation, stage-based fallback, codegen dispatch, collated-predicate fallback, ANSI cast-to-date fix, config aliases)
• Shuffle improvements (spark.comet.shuffle.maxBufferBytes cap, per-writer IPC schema encoding, BatchCoalescer bypass)
• JDK 11 and Spark 3.4 deprecation notice (removal in 1.1.0)
• DataFusion 54.1 and Arrow 58.3 upgrade

Notes for reviewers:

• The post is a draft dated for the planned release date; the changelog link points at the not-yet-generated docs/source/changelog/1.0.0.md.
• No benchmark chart is included in this post.
• Commit/contributor counts (currently ~190 PRs / 22 contributors) are approximate and will be finalized at release time.