Use DeepSeek V4 Flash as the Standard LLM

[RANGROO]

Summary

• replace GLM 4.7 with DeepSeek V4 Flash for the seeded Standard LLM connector
• mark V4 Flash as reasoning-capable so the shared driver explicitly disables DeepSeek's default high-effort thinking mode
• keep Fast, Powerful, and Experimental unchanged; retain GLM 4.7 only as an older-database compatibility fallback

Compatibility

No new migration forces an existing user-selected Standard connector to change.

Validation

• PHP lint passed for debug/db_updates.php, lib/data_functions.php, and ui/quickstart.php
• disposable PostgreSQL fresh-schema load confirmed reasoning_model=true, 750 max tokens, temperature 0.6, JSON/schema enabled, and DeepSeek V4 Flash in both Default Profile and Player Faction Standard slots
• live CHIM shared-driver probe sent reasoning: { exclude: true, enabled: false }, returned valid action JSON in 5.349 seconds, and OpenRouter reported reasoning_tokens: 0
• GitHub PHP Regression Tests passed, including the automatic SQL_ASCII database upgrade
• git diff --check passed

Deployment

Not deployed. The live probe validates the shared connector behavior in CHIM; StobeServer runtime and in-game behavior remain untested.

Related PRs

• Use DeepSeek V4 Flash as the Standard LLM abeiro/HerikaServer#691
• Use DeepSeek V4 Flash as the Standard LLM DialecticServer#65