Slate v2 Target B Richtext Parity And Lists

Supporting plan. For current queue and roadmap truth, see master-roadmap.md.

Goal

Deepen the current richtext family without carrying parity debt:

1. clear block-formatting parity floor
2. add honest list behavior
3. harvest supportable legacy rows immediately
4. keep browser proof ahead of more feature expansion

Scope

1. Finish public mark-helper parity:
   • expanded-selection add/remove
   • leaf host ownership
2. Make block toggles selection-span aware.
3. Add unordered and ordered list toggles with honest unwrap/retoggle behavior.
4. Add indent/outdent only if the current public seam can support it honestly.
5. Harvest supportable legacy rows for each landed seam.
6. Prove all of it through runtime, build, and Chromium example tests.

Progress

• expanded-selection mark add/remove landed in core and Chromium proof
• renderLeaf(...) host ownership landed in slate-react
• next:
  • selection-span-aware block toggles
  • unordered/ordered list toggles
  • immediate legacy harvest for those seams

Findings

• relevant repo learnings mostly reinforce one thing:
  • do not fake selection or wrapper/list semantics with local hacks when the underlying container contract already exists
• current v2 core already recognizes:
  • bulleted-list
  • ordered-list
  • list-item
• legacy Slate richtext uses wrapper-list semantics:
  • list format toggles set selected blocks to list-item
  • then wrap them in bulleted-list or numbered-list