Fix updater handling for superseded release train typo

[justrach]

Summary

• Keep normal semantic version comparison intact for regular releases.
• Add an updater-specific supersession path so known bad release numbers, starting with 0.2.58181, can update to 0.2.5823 or newer instead of being treated as newer forever.
• Add a focused regression test proving pure SemVer still says 0.2.58181 > 0.2.5823, while updater ordering allows the intended forward upgrade.

Validation

• zig build test -Dtest-filter=compareVersionsForUpdate
• zig build test
• zig build
• ./zig-out/bin/codedb --version
• local installed /opt/homebrew/bin/codedb --version -> codedb 0.2.5823
• local installed /Users/blackfloofie/bin/codedb --version -> codedb 0.2.5823
• codedb update -> already up to date

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: bdfa4a7041

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Move the supersession check to code old binaries execute

This special case is added only to the new updater binary, but the affected installed binary is 0.2.58181 and therefore will not contain this table or compareVersionsForUpdate; src/release_info.zig in this commit is still 0.2.5823, so the only shipped binary that can execute this branch is already on the fixed train. Users actually stuck on 0.2.58181 will continue to run the old compareVersions logic and refuse v0.2.5823 via codedb update, so the compatibility shim needs to live in a path the old binary executes, such as the installer/manifest flow, or be delivered by republishing that bad version.

Useful? React with 👍 / 👎.

[github-actions]

Benchmark Regression Report

Thresholds: 10.00% and 50,000 ns absolute delta

NOISE means the percentage threshold was exceeded, but the absolute delta was too small to fail CI.

Tool	Base (ns)	Head (ns)	Delta	Abs Delta (ns)	Status
codedb_bundle	84196	81829	-2.81%	-2367	OK
codedb_changes	9831	9796	-0.36%	-35	OK
codedb_context	935453	884664	-5.43%	-50789	OK
codedb_deps	232	217	-6.47%	-15	OK
codedb_edit	41357	35683	-13.72%	-5674	OK
codedb_find	9150	8840	-3.39%	-310	OK
codedb_hot	24129	25013	+3.66%	+884	OK
codedb_outline	27815	30265	+8.81%	+2450	OK
codedb_read	13992	16540	+18.21%	+2548	NOISE
codedb_search	20426	21098	+3.29%	+672	OK
codedb_snapshot	57419	48610	-15.34%	-8809	OK
codedb_status	8709	8645	-0.73%	-64	OK
codedb_symbol	17896	18620	+4.05%	+724	OK
codedb_tree	35761	34965	-2.23%	-796	OK
codedb_word	10620	11071	+4.25%	+451	OK