fix: migrate _check_types from typeguard to pydantic validate_call

[31puneet]

Overview

Relates to #484

This PR migrates the _check_types runtime parameter validation from typeguard.check_type() to Pydantic v2's validate_call(), as proposed in the issue.

What this PR does

• Replaces the typeguard-based type checking logic in _check_types with Pydantic v2's validate_call()
• Uses validate_call(config=ConfigDict(strict=True, arbitrary_types_allowed=True), validate_return=False) to validate input types without coercing arguments
• Adds pydantic >= 2.0.0 as a dependency
• Preserves existing custom error message format (TypeError with parameter name, value, expected/actual types)
• Replaces silent except Exception: pass with warnings.warn() for unsupported type annotations

Impact

• No changes to public API — all existing @_check_types decorators work unchanged
• typeguard dependency retained for now — still used by the pytest plugin and suppress_type_checks in tests
• Adding Field()-based numeric constraints can be addressed in a follow-up PR if desired

[31puneet]

Hi @jonbrenas
Requesting a review for this PR
Issue #484 also mentions using Pydantic's Field (e.g. Field(ge=0, le=1)) for numerical constraints. Since this requires updating many methods would you prefer me to make those changes or should we keep this PR focused on the validate_call()

[31puneet]

Hi @jonbrenas Could you take a look on this

[31puneet]

Hi @jonbrenas PTAL