chore: add codespell pre-commit hook to catch typos

[anxkhn]

Please ensure you have read the contribution guide before creating a pull request.

Link to Issue or Description of Change

2. Or, if no issue exists, describe the change:

Problem:

There is no automated spell checker in the pre-commit config or CI. The
.pre-commit-config.yaml runs ruff, isort, pyink, addlicense, and mdformat, but
nothing catches misspellings, so typos land in source, docstrings, samples, and
comments and are only fixed later by occasional one-off reader-submitted PRs.

Solution:

Add a codespell hook so typos
are caught automatically, both locally on commit and in CI.

• .pre-commit-config.yaml: add the codespell hook (rev: v2.4.2) with
  additional_dependencies: [tomli] so it can read its config from
  pyproject.toml. The existing lint CI job already runs
  pre-commit/action with no extra args, so the new hook becomes a CI gate
  automatically. No separate workflow job is added, keeping the change minimal.
• pyproject.toml: add a [tool.codespell] table with a small, tuned
  ignore-words-list (genuine false positives only: the ROUGE metric, a couple
  of test-fixture substrings, a local variable name, and intentional
  hyphenation) and a skip list for generated/data files (CHANGELOG, lockfiles,
  notebooks, JSON fixtures, bundled JS/source maps, and the vendored CLI browser
  bundle).
• Fix the existing typos codespell currently flags so the tree stays green
  honestly (typos are fixed, not buried in the ignore list): Initiaze ->
  Initialize, build-in -> built-in, relavant -> relevant, anwer ->
  answer (samples); Coverts -> Converts, exsisting -> existing,
  ouput -> output, incase -> in case, retreving -> retrieving
  (source); simplfy -> simplify, successfull -> successful (test
  comments).

Testing Plan

Unit Tests:

• I have added or updated unit tests for my change.
• All unit tests pass locally.

This is a tooling and typo change (no runtime behavior changes), so no new unit
tests are added. The edits are limited to comments, docstrings, sample text, and
one JSON schema description string; no test asserts on the changed strings.

Verification run locally:

• pre-commit run codespell --all-files -> Passed (exit 0).
• codespell (v2.4.2) bare walk, auto-reading pyproject.toml -> exit 0.
• Full pre-commit suite over the change -> all hooks Passed.
• Targeted pytest on the files with edited comments:
  • tests/unittests/a2a/executor/test_a2a_agent_executor.py -> 25 passed.
  • tests/unittests/integrations/bigquery/test_bigquery_query_tool.py ->
    92 passed.
  • tests/unittests/tools/application_integration_tool/ (covers the edited
    connections_client.py schema description) -> 78 passed.

Manual End-to-End (E2E) Tests:

Not applicable: no user-facing runtime or UI behavior changes. The hook can be
exercised with pre-commit run codespell --all-files.

Checklist

• I have read the CONTRIBUTING.md document.
• I have performed a self-review of my own code.
• I have commented my code, particularly in hard-to-understand areas.
• I have added tests that prove my fix is effective or that my feature works.
• New and existing unit tests pass locally with my changes.
• I have manually tested my changes end-to-end.
• Any dependent changes have been merged and published in downstream modules.

Additional context

The ignore list is intentionally minimal so codespell keeps catching real typos
(verified it still flags things like teh and recieve); only genuine false
positives are suppressed. Happy to drop the source/sample typo fixes into a
separate PR, or split the tool from the fixes, if the maintainers prefer.