2 Commits

Author	SHA1	Message	Date
jgrusewski	609f533abc	feat: implement all 15 fxt CLI commands with real gRPC calls ... - 13 commands with full gRPC implementations: service, train, tune, model, trade, broker, agent, data, risk, config, cluster, auth, backtest - Streaming support: train logs --follow, broker executions --follow - --json output on every command via OutputFormat/HumanReadable - Fix web-gateway monitoring URL default (50057 → 50051, API Gateway) - Rewire 7 remaining build.rs to consolidated proto/ root (web-gateway, backtesting_service, training_uploader, 3 test crates, e2e) - Fix web-gateway ml_training.proto new fields (mode, max_epochs, resume) - 75 fxt tests + 139 web-gateway tests, 0 clippy warnings, workspace clean Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-03-03 22:16:35 +01:00
jgrusewski	0f9d756caa	feat: on-demand training dispatch via K8s Jobs with sidecar uploader ... Extend ml_training_service to dispatch GPU training jobs as K8s batch/v1 Jobs, collect results via a Rust sidecar uploader, and support model promotion with operator approval via fxt CLI. - K8s dispatcher creates Jobs on gpu-training pool with native sidecar - training_uploader crate: watches DONE/FAILED marker, uploads to S3, reports completion via ReportJobCompletion gRPC - PromotionManager compares metrics, queues better models for approval - 4 new proto RPCs: ReportJobCompletion, ListPendingPromotions, ApprovePromotion, RejectPromotion - fxt commands: train start, model list/approve/reject - Training binaries write DONE/FAILED markers + metrics.json - Dockerfile, K8s job template, and CI pipeline updated - StartTraining gracefully falls back to in-process when outside K8s - 27 new tests (16 service + 11 promotion), 141 total service tests pass Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>	2026-02-26 12:43:17 +01:00