Add storage format spec and NDJSON migration tool

[Intina47]

Motivation

• Preserve the existing append-only plain-text notebook while enabling a metadata-first format for sync and richer tooling.
• Define a stable NDJSON schema that supports tags, repo metadata, links, and RFC3339 timestamps without breaking legacy notes.
• Provide a simple, reversible migration path from journal.txt to an optional journal.ndjson for metadata-heavy workflows.

Description

• Add docs/storage-format.md describing the current journal.txt format, a proposed journal.ndjson schema (fields: id, text, created_at, updated_at, tags, links, repo, source), compatibility guarantees, and a migration plan.
• Add scripts/migrate-journal-to-ndjson.go, a small Go utility that reads journal.txt, parses timestamps ([YYYY-MM-DD HH:MM]), emits RFC3339 created_at, preserves original text, assigns a stable hex id, and writes one JSON object per line with source provenance (source file and line number).
• The migration tool supports -in, -out, and -force flags and skips blank lines; IDs are generated with cryptographic randomness and fall back to a timestamped string on error.

Testing

• No automated test suite was run against the new files; this change is documentation and a standalone migration tool.
• The migration script was formatted with gofmt -w scripts/migrate-journal-to-ndjson.go to ensure idiomatic formatting.
• Basic static validation was performed by reading the created files (docs/storage-format.md and scripts/migrate-journal-to-ndjson.go) to confirm the intended schema and flags are present.

---

Codex Task

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: cf860ae4ea

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]

Avoid bufio.Scanner token limit for long notes

This uses bufio.NewScanner with the default 64K token limit, so any journal.txt line longer than ~64KB will cause scanner.Err() to return bufio.ErrTooLong and the migration will stop with a partial journal.ndjson. That can happen if users paste large logs or long-form notes into a single line. Consider switching to bufio.Reader with ReadString('\n') or calling scanner.Buffer with a larger limit so long notes migrate cleanly.

Useful? React with 👍 / 👎.