Add 6 new Fireworks models

[jlegewie]

Add pricing for 6 new Fireworks-hosted models. All prices are per 1M tokens (input / cache read / output):

• deepseek-v4-flash: $0.14 / $0.03 / $0.28
• deepseek-v4-pro: $1.74 / $0.14 / $3.48
• glm-5p1 (GLM-5.1): $1.40 / $0.26 / $4.40
• kimi-k2p6 (Kimi K2.6): $0.95 / $0.16 / $4.00
• minimax-m2p7 (MiniMax M2.7): $0.30 / $0.06 / $1.20
• qwen3p6-plus (Qwen3.6 Plus): $0.50 / $0.10 / $3.00

Source: https://fireworks.ai/models?modelTypes=Serverless

Verification

• make build
• make lint
• make test. 451 passed, 2 pre-existing failures in tests/test_cli.py unrelated to this change. Both seem to be CLI table-rendering tests that also fail on main.

---

Summary by cubic

Adds pricing for six new Fireworks serverless models across packages/js, packages/python, and prices/providers/fireworks.yml, and updates the README model count. Enables accurate cost estimates for these models.

• New Features
  • deepseek-v4-flash: $0.14 / $0.03 / $0.28 per 1M tokens (input / cache read / output)
  • deepseek-v4-pro: $1.74 / $0.14 / $3.48
  • glm-5p1 (GLM-5.1): $1.40 / $0.26 / $4.40
  • kimi-k2p6 (Kimi K2.6): $0.95 / $0.16 / $4.00
  • minimax-m2p7 (MiniMax M2.7): $0.30 / $0.06 / $1.20
  • qwen3p6-plus (Qwen3.6 Plus): $0.50 / $0.10 / $3.00

^{Written for commit 21076df. Summary will update on new commits.}

Review in cubic

[cubic-dev-ai]

No issues found across 6 files

_{Re-trigger cubic}