BFE Files

Fixed

• Fix token usage collection for non-streaming gzip responses (Content-Encoding: gzip with Transfer-Encoding: chunked): the compressed body is now decompressed on a parse-only copy (forwarding bytes untouched), the whole-body parse gate accepts chunked non-SSE JSON bodies, and the content-type decoder tolerates media type parameters (e.g. application/json; charset=utf-8) while passing compressed bodies through untouched, so ai_output_tokens / ai_total_tokens / ai_cache_read_tokens and RMB billing reflect the real usage instead of 0 (issue [#1406])
• Fall back to the model's chat-mode price when the (model, mode) price entry is missing but a chat entry exists (product decision: an unpriced mode bills the same as chat), and mark such requests with the new ai_price_miss access-log field (bfe-access-pb >= v0.3.14, field 763) counted by PriceFallbackUsed, so under-priced rows stay visible in reconciliation (issue [#1408])
• Drop the EstimateToken residue (auth-time prompt seeding / response-byte completion guessing) from the shared TokenUsage and suppress the token fields in the access log for non-2xx responses (new usage disposition, UsageDispositionNone): a failed request no longer logs request-body/4 as ai_input_tokens while ai_total_tokens stays 0 (issue [#1409])

Added

• Make the "usage unrecognized and not estimatable" billing bypass observable (issue [#1410]): the request-finish guard keeps its yardstick behavior (billing fields zeroed, no RMB deduction) but now increments the UsageUnrecognized module counter (exported via monitorHandlers, alertable) and logs a warn with cluster/key_id/model and the response_completed/allow_estimate sub-criteria; the client-abort zero-deduction branch is counted too (counter-only). The access log gains ai_usage_source (bfe-access-pb >= v0.3.15, field 764): parsed (final usage from the upstream response, including zero-output final events) / estimated (approved EstimateToken fallback) / unrecognized (billing skipped, the reported token fields are auth-time seeding residue), orthogonal to the non-2xx suppression, so billing bypasses become directly queryable (ai_usage_source='unrecognized') instead of requiring out-of-band reconciliation. SC03 TC-24 covers the three-way distinction end to end