updated about logs (#5)

updated about scoring evalution (#3)

Harden task score bounds across all inference edge cases

Update inference and environment configuration

Refactor score handling for strict interval compliance

Change SCORE_EPSILON and modify log_end function

Adjust SCORE_EPSILON for precision in scoring

Update SCORE_EPSILON for precision in scoring

Enforce strict (0,1) task scores in exported results

Fix score bounds to strict open interval

Fix score range validation and update deployment config

Harden inference action sanitization limits

Add inference log file output and rotation options

MedCodeRL v1.0 - Full OpenEnv spec compliance, 90 cases, shaped rewards, structured logging

added all files