test: added tests for PromQL to match queries to L3 IR - #14

Restructure the flat 19-section developer guide into the three-part
structure the doc owner asked for: Part 1 - Code Architecture, Part 2
- Interfaces and Definitions, Part 3 - How to Add X, Y, Z (each ending
in how to verify). Content is moved, not rewritten:

Part 1 (Mental model first, per doc-owner follow-up, then a new
whole-PR architecture diagram, then "How the current pieces fit
together"):
- old #1 Mental model -> Part 1 #1
- new: whole-PR architecture diagram (TargetSubDAG's two entry points
  through ReplacementStrategy, PlanSpace/cost_sorted, explanation.rs,
  to a downstream consumer) -> Part 1 #2
- old #3 How the current pieces fit together -> Part 1 #3

Part 2:
- old Terminology's "Implementation" definition merged into the
  Glossary as one more entry (### Implementation), next to
  ReplacementStrategy
- old #2 Glossary -> Part 2 #1 (plus the merged Implementation entry
  and old #10 Matcher, retitled to match glossary-entry style)
- old #10 Matcher (implementation.rs) -> ### Matcher inside the
  Glossary; implementation.rs no longer exists, so the stale title
  is fixed
- old #19's definitional content (ReplacementExplanation/
  ExplanationKind shapes, node_hash, why there's no ExplanationRule
  trait, location-text ownership) -> Part 2 #2

Part 3:
- old #4, #5, #6, #7, #13, #14 -> Part 3 #1, Adding a new
  ReplacementStrategy (ending in Testing a new strategy)
- old #8, #9, #15 -> Part 3 #2, Adding or customizing a CostModel
  (ending in Testing a new cost model)
- old #12 -> Part 3 #3, Adding a new sketch algorithm, with its
  stale implementation.rs/binder references fixed to replacement.rs/
  construct_summary vocabulary, plus a new "Verifying a new sketch
  algorithm" close grounded in the existing coverage-matrix tests
- old #11, #16, #17, #18 -> Part 3 #4-#7 (capstone + closing
  reference material); #18's extension-map table's implementation.rs
  row fixed to replacement.rs
- old #19's "Using it"/"Adding a new kind" content -> Part 3 #8,
  Using and extending explanation.rs

cargo build --workspace --all-targets is clean (docs-only change).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

…t_graph

Found via manual testing against real corpus queries (not the toy
examples used so far): a single, standalone STDDEV_POP aggregate
crashed dag_export --post-asap with a stack overflow, no sharing or
rewriting involved at all.

Root cause: STDDEV_POP (like AVG) dispatches to
Implementation::PassThrough with no alternative strategy of its own,
so SketchAlgorithmStrategy's only candidate for it is
keep_pre_asap()'s conservative fallback -- Replacement::Summary
wrapping the entire target as SummaryExpr::KeepPreAsap. run_post_asap
was treating this like any other winner and handing it to
export_post_asap, whose find_winner re-checks every node inside a
spliced KeepPreAsap payload by design (so a target nested underneath
one still gets found) -- but this payload structurally is the
enclosing target, so the fresh check finds the identical winner again,
unconditionally, forever.

This is the exact "no-op candidate" concept explanation.rs's own
sketch_finding_reason already excludes from being reported as a
finding. run_post_asap's own winner selection just wasn't applying
that same filter. Fixed by skipping a candidate whose
Replacement::Summary is a bare top-level KeepPreAsap when building
winners, with a regression test exercising the exact repro.

Also verified against several real corpus queries end to end
(previously only exercised with hand-written toy SQL): PR-PKT/FT-PKT
(#44/#49 in synthetic_packet_trace_queries.sql) now correctly show a
genuine Rollup relationship, and P-LEN-AVG (#14) shows the real
avg -> sum/count rewrite -- both screenshotted for the PR description.

Verified: cargo test --workspace, fmt, and clippy all clean.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>