Fix regex handling of nested backticks in code block extraction

[chigwell]

User Story
As a maintainer,
I want to correct the test expectations for nested code block extraction
so that our test suite accurately validates proper handling of inner backticks.

Background
The current regex pattern in mdextractor/__init__.py fails to handle nested backticks, splitting the outer block at the first inner triple backticks. This causes the test test_nested_code_blocks in tests/test_mdextractor.py to incorrectly validate ["Outer", "end"] instead of the full "Outer inner end" string. This creates a false positive, masking a critical flaw in code block extraction logic that could impact users relying on nested Markdown structures.

Acceptance Criteria

• Update the regex pattern in mdextractor/__init__.py to properly capture nested backticks without premature block termination.
• Modify the test_nested_code_blocks test case in tests/test_mdextractor.py to expect ["Outer ```inner``` end"].
• Add a validation step to ensure the regex handles multiple levels of nesting (e.g., outer middle ```inner``````).
• Confirm all existing tests pass after changes, including edge cases like single-line and malformed blocks.
• Verify extracted blocks retain original whitespace formatting except for leading/trailing whitespace stripping.