Enhanced testing of minifloat formats

[nickfraser]

Reason for this PR

An enhanced test suite of minifloat formats to address #1126.

Changes Made in this PR

The enhanced test suite generates 10000 random valid minifloat formats / values and checks that Brevitas produces the identical output. Float64 is used for arithmetic to avoid rounding errors in log2, exp2 computations.

Testing Summary

Tests run and pass locally, e.g.:

pytest --hypothesis-verbosity=verbose --hypothesis-show-statistics -n logical tests/brevitas/core/test_float_quant.py::test_valid_float_values
tests/brevitas/core/test_float_quant.py::test_valid_float_values
[gw0] [100%] PASSED tests/brevitas/core/test_float_quant.py::test_valid_float_values
tests/brevitas/core/test_float_quant.py::test_valid_float_values:

  - during generate phase (25.89 seconds):
    - Typical runtimes: ~ 1-2 ms, of which < 1ms in data generation
    - 10000 passing examples, 0 failing examples, 0 invalid examples

  - Stopped because settings.max_examples=10000

Risk Highlight

Only tests added. No change in brevitas code itself. Zero risks IMO.

Checklist

• Code comments added to any hard-to-understand areas, if applicable.
• Changes generate no new warnings.
• Updated any relevant tests, if applicable.
• No conflicts with destination dev branch.
• I reviewed my own code changes.
• Initial CI/CD passing.
• 1+ reviews given, and any review issues addressed and approved.
• Post-review full CI/CD passing.

[nickfraser]

@Giuseppe5, any thoughts on a good value here? Maybe 1000 is a good compromise between confidence and test time...