[Bug] Token Limit Exceeded When Reading Large Log Files

If we get a million tokens, we should be able to load files with less than 100,000 tokens.I understand why you're trying to prevent flooding, but this should probably be configurable or at least scale depending on model size.

Bug Description
sonnet-1M: ● Read(2025-09-26-billing.log.part00)
⎿  Error: File content (72919 tokens) exceeds maximum allowed tokens (25000). Please use offset and limit parameters
to read specific portions of the file, or use the GrepTool to search for specific content.

● Read(2025-09-26-billing.log.part01)
⎿  Error: File content (72521 tokens) exceeds maximum allowed tokens (25000). Please use offset and limit parameters
to read specific portions of the file, or use the GrepTool to search for specific content.

● Read(2025-09-26-billing.log.part02)
⎿  Error: File content (65461 tokens) exceeds maximum allowed tokens (25000). Please use offset and limit parameters
to read specific portions of the file, or use the GrepTool to search for specific content.

Environment Info

• Platform: linux
• Terminal: vte-based
• Version: 1.0.127
• Feedback ID: eb939500-9d68-4717-b8d7-5b5c2d9f4c3f