feat(offset): implement LogQL offset directive — shift eval window for range and instant queries

[szibis]

Summary

• Implements the offset directive gap tracked in KNOWN_ISSUES.md: queries like rate({app="nginx"}[5m] offset 1h) previously silently stripped the offset; they now correctly shift the evaluation window
• Adds extractLogQLOffset() to parse and strip the offset clause from LogQL; multiple distinct offsets in the same query return HTTP 400
• handleQueryRange shifts start/end backward by the offset duration before dispatching to VictoriaLogs; handleQuery shifts time (defaulting to now when omitted)
• Offset extraction runs before preferWorkingParser so the parser probe targets the shifted historical window

Test Plan

• go test ./internal/proxy/... -run 'Offset|offset' -v — 18 offset unit + integration tests pass
• go test ./... — 2518 tests, no regressions
• Manual: query with offset 1h returns data from 1 hour ago (not current window)
• Manual: query with two different offsets (offset 1h + offset 2h) returns HTTP 400
• Manual: instant query (/loki/api/v1/query) with offset 30m and no time param shifts correctly to now-30m

[github-actions]

PR Quality Report

Compared against base branch main.

Coverage and tests

Signal	Base	PR	Delta
Test count	2507	2524	17
Coverage	87.2%	87.2%	0.0% (stable)

Compatibility

Track	Base	PR	Delta
Loki API	100.0%	11/11 (100.0%)	0.0% (stable)
Logs Drilldown	100.0%	17/17 (100.0%)	0.0% (stable)
VictoriaLogs	100.0%	11/11 (100.0%)	0.0% (stable)

Performance smoke

Lower CPU cost (ns/op) is better. Lower benchmark memory cost (B/op, allocs/op) is better. Higher throughput is better. Lower load-test memory growth is better. Benchmark rows are medians from repeated samples.

Signal	Base	PR	Delta
QueryRange cache-hit CPU cost	1778.0 ns/op	1770.0 ns/op	-0.4% (stable)
QueryRange cache-hit memory	200.0 B/op	200.0 B/op	0.0% (stable)
QueryRange cache-hit allocations	7.0 allocs/op	7.0 allocs/op	0.0% (stable)
QueryRange cache-bypass CPU cost	2021.0 ns/op	2024.0 ns/op	+0.1% (stable)
QueryRange cache-bypass memory	286.0 B/op	285.0 B/op	-0.3% (stable)
QueryRange cache-bypass allocations	7.0 allocs/op	7.0 allocs/op	0.0% (stable)
Labels cache-hit CPU cost	685.0 ns/op	685.5 ns/op	+0.1% (stable)
Labels cache-hit memory	48.0 B/op	48.0 B/op	0.0% (stable)
Labels cache-hit allocations	3.0 allocs/op	3.0 allocs/op	0.0% (stable)
Labels cache-bypass CPU cost	825.9 ns/op	811.5 ns/op	-1.7% (stable)
Labels cache-bypass memory	53.0 B/op	53.0 B/op	0.0% (stable)
Labels cache-bypass allocations	3.0 allocs/op	3.0 allocs/op	0.0% (stable)

State

• Coverage, compatibility, and sampled performance are reported here from the same PR workflow.
• This is a delta report, not a release gate by itself. Required checks still decide merge safety.
• Performance is a smoke comparison, not a full benchmark lab run.
• Delta states use the same noise guards as the quality gate (percent + absolute + low-baseline checks), so report labels match merge-gate behavior.