[FEATURE]: Restore DeepSeek V4 Flash quota on Go to pre-August levels #46877

Feature hasn't been suggested before.

• I have verified this feature I'm about to request hasn't been suggested before.

Describe the enhancement you want to request

What would you like to see added or improved?

Restore the DeepSeek V4 Flash quota on OpenCode Go to its pre-August level
(31,650 requests/5h, $60 monthly cap), or at least a meaningful increase over
the current 7,600 requests/5h / $30 cap.

Why is this important?

DeepSeek V4 Flash was the workhorse of the Go subscription. At the previous
quota it was effectively free for real workloads: I used it as my primary
coding model for a full month (late July to early August 2026), running long
agentic sessions daily, and never came close to exhausting the 5-hour, weekly,
or monthly limits. The experience was excellent, and it was the main reason I
subscribed to Go.

The current quota (7,600 req/5h, $30 cap) is a ~4.2x reduction, and combined
with the new peak/off-peak pricing it makes DeepSeek V4 Flash no longer
viable as a primary model for heavy agentic use. The docs state "limits may
change as we learn from early usage and feedback" — this is feedback from a
paying user who used the previous limits heavily and found them well-calibrated.

Additional context

• I understand the peak/off-peak pricing reflects DeepSeek's own price changes
  and is not something Go can fully control.
• I'm not asking for free access beyond the subscription: I'm asking for the
  quota to reflect the value the subscription used to deliver.
• Related: [FEATURE] Optional quota multipliers (x2 / x4 / x8 / x16) for Go subscription with volume discount #43194 (quota multipliers) would be a complementary path for users
  who want even more headroom and are willing to pay for it.