Improve JSON and Pydantic negative tests with more `ParserException` info

[hudson-ai]

This PR adds additional info to ParserExceptions to facilitate more explicit and stringent test conditions when a string is expected not to match a grammar.

Changes to _parser.py:

• Adds consumed_bytes field to ParserException to communicate to the user which bytes were consumed before (and including) the failure.
• Includes allowed_bytes and consumed_bytes in the ParserException when applicable

Changes to tests:

• Rewrite "negative" tests for test_json.py and test_pydantic.py (the only two test files that used ParserExceptions to make asserts) to explicitly assert the expected values of allowed_bytes and consumed_bytes from the exception.
• Refactor common "check_match_failure" to tests/utils.py

[hudson-ai]

@riedgar-ms request for review

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests :white_check_mark:

Project coverage is 59.85%. Comparing base (cb263a8) to head (0273662).

:exclamation: Your organization needs to install the Codecov GitHub app to enable full functionality.

Additional details and impacted files

```diff @@ Coverage Diff @@ ## main #864 +/- ## ========================================== + Coverage 56.91% 59.85% +2.94% ========================================== Files 63 63 Lines 4570 4571 +1 ========================================== + Hits 2601 2736 +135 + Misses 1969 1835 -134 ```

:umbrella: View full report in Codecov by Sentry.
:loudspeaker: Have feedback on the report? Share it here.

[Harsha-Nori]

LGTM