liturgy: Section type + HTML parse (fixtures)

Ports parse_sections, html_to_lines, extract_reference from ewangelia.py.
Parse(html) prefers the tabnowy0all lectionary tab (falls back to
tabstary0all) and errors loudly if the tab is present-but-empty or
absent. ExtractCitation(heading) pulls the parenthetical citation.

Section gains a PartID field (Addendum A): set from the heading prefix
(1. czytanie -> pierwsze_czytanie, a second same-day 1. czytanie ->
drugie_czytanie, Psalm -> psalm, Aklamacja -> aklamacja, Ewangelia ->
ewangelia; anything else -> "").

Adds testdata/2026-07-22.html (split-feast day, two 1. czytanie
sections) and testdata/2026-06-22.html (normal day), copied from
~/.cache/daily-reading/.

1 files changed, 12 insertions, 0 deletions

diff --git a/internal/liturgy/section.go b/internal/liturgy/section.go
new file mode 100644
index 0000000..498b201
--- /dev/null
+++ b/internal/liturgy/section.go
@@ -0,0 +1,12 @@
+// Package liturgy parses the daily Catholic liturgy readings page into
+// structured sections (1st reading, psalm, acclamation, gospel).
+package liturgy
+
+// Section is one reading section of the liturgy page (e.g. "1. czytanie",
+// "Psalm", "Aklamacja", "Ewangelia"): its heading, optional subtitle, the
+// citation extracted from the heading, a stable identifier for which liturgical
+// part it is, and its body split into paragraphs of text lines.
+type Section struct {
+	Heading, Subtitle, Citation, PartID string
+	Paragraphs                          [][]string
+}