ty: un-ignore invalid-return-type rule

[SirHumza]

Describe your change

• Add an algorithm?
• Fix a bug or typo in an existing algorithm?
• Add or change doctests? -- Note: Please avoid changing both code and tests in a single pull request.
• Documentation change?

Contributes to #15187.

Removes rules.invalid-return-type = "ignore" from [tool.ty]. On a synced Python 3.15 (regular GIL) environment the rule now reports zero diagnostics.

Files fixed:

• data_structures/arrays/dynamic_array.py -- annotated self.array as list[int | None] and narrowed the value in get (the bounds check already guarantees a filled slot).
• graphs/bi_directional_dijkstra.py -- bidirectional_dij can return the np.inf sentinel / fractional path cost, so the return annotation is float, not int.
• graphs/hopcroft_karp.py -- pair_left stores T | object (the _NIL sentinel); the comprehension filters _NIL out, so the result is cast to dict[T, T].
• computer_vision/mosaic_augmentation.py -- the returned image is an np.ndarray, not a list; annotated the return as tuple[np.ndarray, list, str] and typed path_list as list[str].
• machine_learning/mfcc.py -- mel_to_freq is called with both a scalar and an array; added overloads so the array call returns an np.ndarray (matching its tuple[np.ndarray, np.ndarray] return annotation).
• networking_flow/ford_fulkerson.py -- the graph is untyped (list), so the accumulator can be int | float; widened the return annotation accordingly.
• other/dancing_links.py -- _choose_column can return None when no columns remain (already handled by the caller); annotated ColumnNode | None.
• web_programming/random_anime_character.py -- narrowed the parsed <meta> content to str before returning the tuple[str, str, str].

Verification (uv sync --python 3.15, then uv run --python 3.15 --with ty ty check):

• No invalid-return-type diagnostics remain and the total diagnostic count does not increase.
• doctests pass for the touched modules (dynamic_array.py, bi_directional_dijkstra.py, dancing_links.py, mfcc.py, hopcroft_karp.py, ford_fulkerson.py).
• ruff check and ruff format --check clean on the touched files.

Checklist

• I have read CONTRIBUTING.md.
• This pull request is all my own work -- I have not plagiarized.
• I know that pull requests will not be merged if they fail the automated tests.
• This PR only changes one algorithm file. To ease review, please open separate PRs for separate algorithms.
• All new Python files are placed inside an existing directory.
• All filenames are in all lowercase characters with no spaces or dashes.
• All functions and variable names follow Python naming conventions.
• All function parameters and return values are annotated with Python type hints.
• All functions have doctests that pass the automated testing.
• All new algorithms include at least one URL that points to Wikipedia or another similar explanation.
• If this pull request resolves one or more open issues, then the description above includes the issue number(s) with a closing keyword: "Fixes #ISSUE-NUMBER".