hotfix/code_update_after_fix: The code is properly updated after fixing the code

[Maths-A]

No description provided.

[Copilot]

Pull request overview

Adds client-side normalization and a shared helper to consistently apply AI-generated/edited code updates to notebook cells in the Plainbook web UI.

Changes:

• Introduces normalizeSource() to coerce cell source content into a consistent string format.
• Adds applyGeneratedCode() to centralize updating a cell’s source, clearing outputs, and clearing validation state.
• Uses these helpers across code generation, test-code generation, and code-save flows to keep client state consistent.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

When regenerating unit test sub-cell code for roles other than target, the code and outputs are replaced but any existing metadata.validation for that sub-cell is left intact. This can show stale validation results for code that no longer matches the cell content. Consider clearing test.cells[role].metadata.validation (and any related timestamp field, if present) at the same time you reset source/outputs (mirroring applyGeneratedCode).

Suggested change

	test.cells[role].outputs = [];
	test.cells[role].outputs = [];
	if (test.cells[role].metadata) {
	delete test.cells[role].metadata.validation;
	delete test.cells[role].metadata.validation_timestamp;
	delete test.cells[role].metadata.validationTimestamp;
	}