feat: add multiple longest common subsequence (MLCS-APP)

[Gaspardcode]

Description

Adds multiple_longest_common_subsequence in src/string: finds a longest common subsequence of any number of strings (the classic DP only handles two efficiently, MLCS is NP-hard in general).

It implements MLCS-APP from A Fast Heuristic Search Algorithm for Finding the Longest Common Subsequence of Multiple Strings (Wang, Pan, Shang, Korkin, AAAI 2010) — paper, an A* variant over the matches of the strings:

• h(p) = min of the suffix tables M(i,i+1)[p_i][p_i+1] (only the d − 1 consecutive tables, as in the paper), an upper bound of the remaining LCS
• f(p) = g(p) + h(p), points ordered by f then h in a BinaryHeap
• the successors are found in O(d) per letter with the MT lookup table
• each layer keeps the points with y − c ≤ f(p) and at most k of them, with c = 20 and k = 2000 (the values used in the paper)

Two details go slightly beyond the pseudocode so that the result is always valid and as long as possible:

• g(q) is only updated through a longer path (as in MLCS-A*), so the parent chain of a point never gets shorter
• a point is returned only when no remaining point has a higher f (the condition of Theorem 1), and a point without any successor also ends a subsequence, since h is only an upper bound (e.g. ["BB", "BBC", "BC"] has h = 1 at the single match but no successor)

Tests check the length of the result and that it is a subsequence of every input, since several LCS may exist. Edge cases: no string, a single string, empty strings, no common letter, unicode.
I also cross-checked it locally against an exact DP on 21,300 random inputs (2 to 4 strings, alphabets of 3 and 4 letters): the result was optimal in every case.

This supersedes #861 (closed as stale); its review comments are addressed (Option<usize> instead of a sentinel, Context methods, BinaryHeap, link to doc/paper.pdf).

Type of change

• New feature (non-breaking change which adds functionality)

Checklist:

• I ran cargo clippy --all -- -D warnings just before my last commit and fixed any issue that was found.
• I ran cargo fmt just before my last commit.
• I ran cargo test just before my last commit and all tests passed.
• I added my algorithm to the corresponding mod.rs file within its own folder, and in any parent folder(s).
• I added my algorithm to DIRECTORY.md with the correct link.
• I checked COUNTRIBUTING.md and my code follows its guidelines.

The whole MLCS test module runs in ~20 ms.

🤖 Generated with Claude Code

[codecov-commenter]

Codecov Report

❌ Patch coverage is 99.07407% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 95.88%. Comparing base (2345c66) to head (f5ead9d).

Files with missing lines	Patch %	Lines
src/string/multiple_longest_common_subsequence.rs	99.07%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #1070      +/-   ##
==========================================
- Coverage   95.89%   95.88%   -0.02%     
==========================================
  Files         396      397       +1     
  Lines       30440    30656     +216     
==========================================
+ Hits        29191    29395     +204     
- Misses       1249     1261      +12     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.