[CCOR-13507] LLM Chat complete task that uses AI Prompt in a saved workflow should show it in prompt template and not text in custom instructions

[ajlai-orkes]

Pull Request type

• Bugfix
• Feature
• Refactoring (no functional changes, no api changes)
• Build related changes
• WHOSUSING.md
• Other (please describe):

NOTE: Please remember to run ./gradlew spotlessApply to fix any format violations.

Changes in this PR

https://orkes.atlassian.net/browse/CCOR-13507

Summary

• Reloading an LLM Chat Complete task that used a saved AI Prompt was treating the prompt name as custom instructions because the picker only counted a value as a template if it was already in the fetched option list (which is empty until /prompts returns).
• Mode is now driven by allowRawPrompts (same as the server): a non-raw instructions value stays selected in Prompt Template; custom text stays in the textarea.
• The form machine loads prompt names on startup instead of waiting for focus, and the picker derives expansion from that state instead of useEffect.

https://www.loom.com/share/6b9d200da41440a9b93728828fe9719d