v1.0.0-rc.178 (#88)

* Rugnux: Processes serial crystallography data from XFELs - SACLA (MPCCD), PAL-XFEL and European XFEL (CXI), LCLS (XTC) and SwissFEL Alvra (JUNGFRAU with a CrystFEL module geometry) - with several files or Jungfraujoch master files merged in one run (`rugnux *_master.h5`).
* Rugnux: Stills processing takes a known unit cell and space group as a starting point, finds the beam centre (and, with a cell, the detector distance) from the images, detects the saturation level in other facilities' files and refines the placement of individual detector panels.
* Rugnux: The stills report gives R_split and CC1/2 from frame half-sets, comparable with CrystFEL.
* jfjoch_viewer: Opens the same XFEL file formats.
* Validation: A stills battery checks Rugnux against deposited structures from SACLA, PAL-XFEL, LCLS, European XFEL and SwissFEL.
* Rugnux: More reliable convergence of rotation scaling, and live molecular replacement starts as soon as the model is uploaded.
* jfjoch_broker: Initialize clears the previous dataset's bearer tokens, image buffer and live Rugnux status, and a slow CUDA JIT cache is reported at start.

Reviewed-on: #88
Co-authored-by: Filip Leonarski <filip.leonarski@psi.ch>

tests/MergeScaleTest.cpp

@@ -863,3 +863,54 @@ TEST_CASE("SublatticeRelabel_SignalPicksTheAxisThatWasDoubled") {
         CHECK((ScatteringVector(outcomes[0].latt, r) - ScatteringVector(before.latt, b)).Length() < 1e-5f);
     }
 }
+
+// R_split's half-sets keep every frame whole: each frame measures the same reflections at its own
+// intensity, so every reflection's two half-set means are the same two numbers. R_split is then
+// (1/sqrt 2) |I_1 - I_2| / ((I_1 + I_2) / 2) in every shell, and the frame-split CC1/2 is undefined
+// (the half-set means do not vary) - the per-reflection balanced split, which the processing decides on,
+// is left as it was.
+TEST_CASE("MergeOnTheFly_RsplitKeepsEveryFrameWhole") {
+    DiffractionExperiment x;
+    x.SetSpaceGroup(TestSpaceGroup());
+
+    std::vector<IntegrationOutcome> outcomes(64);
+    for (size_t f = 0; f < outcomes.size(); ++f) {
+        for (int h = 1; h <= 4; ++h)
+            for (int k = 1; k <= 4; ++k) {
+                Reflection r{};
+                r.h = h;
+                r.k = k;
+                r.l = 1;
+                r.d = 20.0f / std::sqrt(static_cast<float>(h * h + k * k + 1));
+                r.I = 100.0f + static_cast<float>(f);
+                r.sigma = 10.0f;
+                r.prescaling_corr = 1.0f;
+                r.partiality = 1.0f;
+                r.image_scale_corr = 1.0f;
+                outcomes[f].reflections.push_back(r);
+            }
+    }
+
+    MergeOnTheFly merge(x);
+    merge.AddImages(outcomes);
+    const auto merged = merge.ExportReflections();
+    REQUIRE(merged.size() == 16);
+
+    const float a = merged[0].I_split[0], b = merged[0].I_split[1];
+    REQUIRE(std::isfinite(a));
+    REQUIRE(std::isfinite(b));
+    CHECK(a != b);
+    for (const auto &m: merged) {
+        CHECK(m.I_split[0] == a);
+        CHECK(m.I_split[1] == b);
+    }
+
+    const auto stats = merge.MergeStats(merged, outcomes);
+    const double expected = std::fabs(a - b) / (0.5 * (a + b)) / std::sqrt(2.0);
+    CHECK(stats.overall.n_split_pairs == 16);
+    CHECK(stats.overall.r_split == Catch::Approx(expected).epsilon(1e-5));
+    for (const auto &sh: stats.shells)
+        if (sh.n_split_pairs > 0)
+            CHECK(sh.r_split == Catch::Approx(expected).epsilon(1e-5));
+    CHECK_FALSE(SerialShellTable(stats).empty());
+}