chore(deps): update zrip requirement from 0.7 to 0.8

[dependabot]

Updates the requirements on zrip to permit the latest version.

Changelog

Sourced from zrip's changelog.

[0.8.1]

Changed

• Reused Huffman decode tables and MaybeUninit FSE sequence table storage to remove per-block decode allocations.
• Specialized decoder no-history execution and sequence table reads to reduce decode hot-loop work.
• Specialized DFast MLS4, Fast L1 hash-log, and rare Fast MLS7 encode hot paths.
• Preserved copied MaybeUninit sequence entries for the default path while borrowing sequence table entries in paranoid builds, recovering paranoid decode throughput.
• Refreshed x86_64 benchmark charts after rebenchmarking zrip and zrip paranoid with performance governor and turbo off.
• JSR package @paddor/zrip bumped to 0.5.1 for the current Rust codec.

[0.8.0]

Added

• Level -8 (zrip-only): same parameters as old L-7 but with force_raw_literals, extending the level range from [-7, 4] to [-8, 4]. Old L-7 now gets Huffman-compressed literals while L-8 trades ratio for maximum encode throughput.
• 3-tier dictionary setup in CompressContext adapts hash table initialization to input size: micro inputs use a linear-scan attached match finder with raw literals, small inputs add Huffman from dict tables, large inputs copy the full hash snapshot.
• Decode-side dictionary FSE/Huffman caching: pre-promotes dict tables at construction time so each frame clones promoted tables instead of rebuilding from raw dictionary entries.

Changed

• L2 hash_log raised from 16 (256 KB) to 17 (512 KB) for better ratio.
• huf_worth_trying minimum gain threshold raised from n/256 (~0.4%) to n/32 (~3%) to skip Huffman table builds on marginally compressible blocks.
• Per-level raw literals size ramp skips Huffman for tiny blocks.
• Decode doctests changed to no_run so Miri does not attempt fearless_simd::dispatch! AVX2 paths it cannot emulate.

[0.7.3]

Fixed

• Three UB vulnerabilities found by miri:
  • ReverseBitReader::refill_fast could underflow ptr and read out of bounds on short streams. Now returns early instead of relying on debug_assert!.

... (truncated)

Commits

• 420a874 Merge pull request #39 from paddor/release-0.8.1
• a11e20b Prepare 0.8.1 release
• de54f33 Merge pull request #38 from paddor/codec-perf-specialization
• 91be543 Refresh x86_64 benchmark charts
• 1007ee9 Recover paranoid sequence table decode perf
• a2a389c Specialize Fast L1 hash log
• 5ab3b4b Specialize rare Fast MLS7 path
• a5094aa Specialize DFast MLS4 hot path
• 9bd597c Hash decoded blocks while cache hot
• 3e2aa37 Copy sequence table entries in decode hot loop
• 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)