docs: use colons in place of em dashes in gloss bullets

[lesnik512]

The docs style guide kept em dashes in four places: See-also bullets, term-gloss bullets, table cells and code comments. The shared prose check drafted in modern-python/.github#134 fails on any em or en dash outside code, so modern-di can't adopt it while those exceptions stand.

This replaces the remaining 55 dashes with colons:

• - [Link](target) — gloss becomes - [Link](target): gloss in the See-also lists on the performance, provider and troubleshooting pages.
• - **Term** — gloss and - `file.py` — gloss in AGENTS.md get the same treatment.
• When the gloss started on a continuation line ( — how ...), the colon moves to the end of the link line.
• The en dash in AGENTS.md's "1–2 lines" becomes "one or two lines".

The "Em dashes" section of docs/agents/docs-style.md now says there are no dashes outside code, and AGENTS.md's pointer to it no longer mentions carve-outs. Dashes in fenced code (the migration pages and grpc.md) are left alone, as are headings, anchors and link targets.

Checked with:

• A fence-aware scan for U+2013/U+2014 across README.md, CONTEXT.md, AGENTS.md and docs/ outside docs/agents/, with inline code stripped: 0 hits.
• mkdocs build --strict: clean.
• just test-docs: 218 passed, 27 skipped.

[github-actions]

Benchmark

Details

Benchmark suite	Current: 7f21c67	Previous: 996cf18	Ratio
benchmarks/test_guard_by_type.py::test_g16_resolve_by_type	3334406.9012005804 iter/sec (stddev: 1.6731890822610256e-8)	3398971.0906722094 iter/sec (stddev: 1.1594998695345776e-8)	1.02
benchmarks/test_guard_by_type.py::test_g17_resolve_by_type_large_registry	3327674.401170966 iter/sec (stddev: 1.2860770696495418e-8)	3312988.93090924 iter/sec (stddev: 1.1487165001759097e-8)	1.00
benchmarks/test_guard_cold.py::test_g8_cold_first_resolve	16143.281105650654 iter/sec (stddev: 0.00005019689942117853)	15157.436795676023 iter/sec (stddev: 0.0000552800367174803)	0.94
benchmarks/test_guard_cold.py::test_g8b_cold_first_resolve_cached	12676.70625206963 iter/sec (stddev: 0.00012839400875550204)	12119.638347945027 iter/sec (stddev: 0.00014695476835884408)	0.96
benchmarks/test_guard_concurrency.py::test_g14_concurrent_cached_hit[1]	478.34005751656247 iter/sec (stddev: 0.00003188028560265708)	480.4117649154862 iter/sec (stddev: 0.00005657601867778982)	1.00
benchmarks/test_guard_concurrency.py::test_g14_concurrent_cached_hit[2]	457.99123487218657 iter/sec (stddev: 0.00013077284320863996)	465.5165380259087 iter/sec (stddev: 0.000043030175168246065)	1.02
benchmarks/test_guard_concurrency.py::test_g14_concurrent_cached_hit[4]	422.31740457809786 iter/sec (stddev: 0.00019769058509838919)	444.096702549534 iter/sec (stddev: 0.00004985484604278839)	1.05
benchmarks/test_guard_concurrency.py::test_g15_concurrent_first_resolve[1]	12686.903193503531 iter/sec (stddev: 0.000009899049066435217)	13369.973001569506 iter/sec (stddev: 0.000008366720880480949)	1.05
benchmarks/test_guard_concurrency.py::test_g15_concurrent_first_resolve[2]	6414.131614789361 iter/sec (stddev: 0.000022673108320708026)	6267.410016324864 iter/sec (stddev: 0.000022659937500857765)	0.98
benchmarks/test_guard_concurrency.py::test_g15_concurrent_first_resolve[4]	3289.299030366925 iter/sec (stddev: 0.000054570758275147405)	3322.707499348036 iter/sec (stddev: 0.00004048375177055468)	1.01
benchmarks/test_guard_concurrency.py::test_g15b_concurrent_first_resolve_sibling_children[1]	10213.219685127955 iter/sec (stddev: 0.000011267280334269765)	9994.712796954176 iter/sec (stddev: 0.000009678071802466218)	0.98
benchmarks/test_guard_concurrency.py::test_g15b_concurrent_first_resolve_sibling_children[2]	4122.599796630824 iter/sec (stddev: 0.00002815958937796498)	4167.936005794685 iter/sec (stddev: 0.00006157674202713174)	1.01
benchmarks/test_guard_concurrency.py::test_g15b_concurrent_first_resolve_sibling_children[4]	1969.2300582885796 iter/sec (stddev: 0.00005551557733592266)	1993.2355233848903 iter/sec (stddev: 0.000044461228970224894)	1.01
benchmarks/test_guard_concurrency.py::test_g15c_worker_pool_floor_control[1]	36492.24387782432 iter/sec (stddev: 0.00000557067922295882)	39645.974660995125 iter/sec (stddev: 0.000007233125204690835)	1.09
benchmarks/test_guard_concurrency.py::test_g15c_worker_pool_floor_control[2]	14226.80515044469 iter/sec (stddev: 0.0000142301640977439)	13691.70144831524 iter/sec (stddev: 0.000011074355202792832)	0.96
benchmarks/test_guard_concurrency.py::test_g15c_worker_pool_floor_control[4]	5716.565807399165 iter/sec (stddev: 0.00003428660406704914)	5752.9347517906735 iter/sec (stddev: 0.000030141119359696113)	1.01
benchmarks/test_guard_lifecycle.py::test_g6_build_child_container	1508107.4537118599 iter/sec (stddev: 9.274250992012504e-8)	1601065.348880601 iter/sec (stddev: 3.351404397134636e-8)	1.06
benchmarks/test_guard_lifecycle.py::test_g6b_build_child_container_auto_scope	1685683.991556147 iter/sec (stddev: 3.446069892805093e-8)	1698991.10510454 iter/sec (stddev: 1.7713780602138135e-8)	1.01
benchmarks/test_guard_lifecycle.py::test_g7_request_lifecycle_batch	3109.0694506529294 iter/sec (stddev: 0.00001075674151439705)	3154.7025979758973 iter/sec (stddev: 0.00002583232235798169)	1.01
benchmarks/test_guard_lifecycle.py::test_g7c_event_loop_floor_control	58144.10231999106 iter/sec (stddev: 0.0000018710723234150708)	58215.67659786307 iter/sec (stddev: 0.0000020595849361970636)	1.00
benchmarks/test_guard_lifecycle.py::test_g7b_request_cycle_sync	409645.54681698 iter/sec (stddev: 9.105448957161004e-8)	411538.680366211 iter/sec (stddev: 8.352210387897596e-8)	1.00
benchmarks/test_guard_lifecycle.py::test_g13_teardown_at_scale	66329.76992924025 iter/sec (stddev: 0.000001708678238888406)	65542.96247024715 iter/sec (stddev: 0.0000016247394412795976)	0.99
benchmarks/test_guard_lifecycle.py::test_g13b_teardown_at_scale_async_no_finalizers	596.8354148662207 iter/sec (stddev: 0.0007034387975168722)	592.6892893316399 iter/sec (stddev: 0.0009227163026292208)	0.99
benchmarks/test_guard_resolve.py::test_g1_transient_resolve	1659898.9818743286 iter/sec (stddev: 1.6628854815430255e-7)	2193069.024660947 iter/sec (stddev: 3.110757504734808e-8)	1.32
benchmarks/test_guard_resolve.py::test_g2_cached_resolve	3550908.127004279 iter/sec (stddev: 1.6083891854179598e-8)	3507523.86667632 iter/sec (stddev: 8.278846667384555e-9)	0.99
benchmarks/test_guard_resolve.py::test_g3_deep_chain	752223.6719033258 iter/sec (stddev: 1.5315660782444867e-7)	786886.7216373547 iter/sec (stddev: 4.928512472057435e-8)	1.05
benchmarks/test_guard_resolve.py::test_g4_wide_resolve	429652.7602246477 iter/sec (stddev: 6.720770775357902e-7)	474191.17330059607 iter/sec (stddev: 4.1593229872813204e-7)	1.10
benchmarks/test_guard_resolve.py::test_g5_cross_scope	1915812.1266549237 iter/sec (stddev: 9.483863232628565e-8)	1958485.4933757903 iter/sec (stddev: 2.7805444908867615e-8)	1.02
benchmarks/test_guard_resolve.py::test_g9_context_resolve	1377056.909908989 iter/sec (stddev: 1.6457402221986638e-7)	1423463.1784436998 iter/sec (stddev: 1.4739503375312504e-7)	1.03
benchmarks/test_guard_resolve.py::test_g12_override_active_resolve	835502.5416223381 iter/sec (stddev: 5.974050875853733e-8)	833961.9669145662 iter/sec (stddev: 4.029536057763587e-8)	1.00
benchmarks/test_guard_resolve.py::test_g18_alias_hop	3634779.7975946483 iter/sec (stddev: 9.327210379331227e-9)	3648776.1092819464 iter/sec (stddev: 1.2293149854723347e-8)	1.00
benchmarks/test_guard_validate.py::test_g10_validate_deep_chain	31101.493607804667 iter/sec (stddev: 0.00021484920536539833)	34663.69874541447 iter/sec (stddev: 0.000018809888676094653)	1.11
benchmarks/test_guard_validate.py::test_g11_validate_wide	20678.96956444341 iter/sec (stddev: 0.00002463306434591771)	20923.262532802193 iter/sec (stddev: 0.00002336397928812689)	1.01

This comment was automatically generated by workflow using github-action-benchmark.