Reorganize task categories from opaque a/b to descriptive names

Replace category_a/category_b directories with algorithm, pipeline,
environment, filesystem, and process. Add separate mode field (solve/convert)
to decouple orchestration from capability grouping. Add per-category
summary and questionnaire breakdowns to both terminal report and HTML export.

tasks/pipeline/pipeline_word_freq.toml

@@ -0,0 +1,37 @@
+name = "pipeline_word_freq"
+category = "pipeline"
+mode = "convert"
+description = """
+Read text from stdin. Count the frequency of each word (case-insensitive, only alphabetic characters count as words).
+Print the top 5 most frequent words in descending order of frequency, in the format:
+"count word"
+If two words have the same count, sort them alphabetically.
+If there are fewer than 5 unique words, print all of them.
+"""
+
+bash_source = """
+#!/bin/bash
+tr '[:upper:]' '[:lower:]' | tr -cs '[:alpha:]' '\n' | grep -v '^$' | sort | uniq -c | sort -k1,1rn -k2,2 | head -5 | while read -r count word || [[ -n "$word" ]]; do
+    echo "$count $word"
+done
+"""
+
+[[test_cases]]
+stdin = """The quick brown fox jumps over the lazy dog.
+The dog barked at the fox. The fox ran away."""
+expected_stdout = """5 the
+3 fox
+2 dog
+1 at
+1 away"""
+
+[[test_cases]]
+stdin = "hello hello world"
+expected_stdout = """2 hello
+1 world"""
+
+[[test_cases]]
+stdin = "One one ONE two TWO two Three three three three"
+expected_stdout = """4 three
+3 one
+3 two"""