Deprecate separate title attribute in flows; derive from description

[Copilot]

Flow YAML files required a redundant top-level title attribute even when the description already began with a Markdown heading—unlike flow pages and static-page chunks, which derive their title from markup automatically.

Changes

• FlowDesc.title — changed from required str to str | None = None
• FlowDesc.check_title validator — issues a deprecation warning via the validation context when title is explicitly supplied and extract_title_from_markup(description) would have yielded a result; raises ValueError when neither source provides a title
• FlowDesc.get_title() — new method mirroring ChunkDesc.get_title(): returns title if set, otherwise extracts from description
• models.py and 5 flow templates — updated all flow_desc.title accesses to flow_desc.get_title() / flow_desc.get_title

Example

A flow description that previously required both:

title: "Quiz 1"
description: |
  # Quiz 1
  ...

can now omit title entirely:

description: |
  # Quiz 1
  ...

Keeping both will produce a validation warning pointing authors to remove the redundant attribute.