fix(crash-reporting): pin the empty-ceiling-file term and correct the uptime claim

An empty memory.max reads as Number('') = 0 without the empty-string term, so a
sandbox that stubs /sys/fs/cgroup with zero-length files would ship a 0 MB
ceiling and a false mem-available-cgroup-capped label. No test held that term.

The doc said the main process started 'hours earlier' in all three reports;
2ea53f9c is 42 m 55 s. Replaced with the measured per-report uptimes.

docs/reference/linux-memory-kill-attribution.md

@@ -18,9 +18,10 @@ The kernel OOM killer does not fire with that much headroom. `181e8e36` is two
 single-process kills, not one whole-cgroup kill: the `process_gone_suppressed`
 GPU crumb is at `22:29:32.397Z` and the renderer report at `22:38:37.276Z`, so
 they are not co-timed and nothing links them beyond the host. And in all three
-Orca's **main** process survived and stayed the reporter — `mainProcessStartedAt`
-hours earlier, `processMetricsBrowserCount: 1` in the post-death sample — which
-is not what a whole-cgroup kill leaves behind.
+Orca's **main** process survived and stayed the reporter — already up 43 m
+(`2ea53f9c`), 11 h (`ad185d76`) and 3 h 11 m (`181e8e36`) by
+`mainProcessStartedAt`, and `processMetricsBrowserCount: 1` in the post-death
+sample — which is not what a whole-cgroup kill leaves behind.
 
 That last point is why `systemd-oomd` is a candidate here and not a conclusion:
 it kills the whole cgroup, and the surviving main process argues against it for

src/main/crash-reporting/linux-cgroup-memory-limit.test.ts

@@ -106,6 +106,36 @@ describe('cgroup v2 memory directory resolution', () => {
     })
   })
 
+  it('does not turn a zero-length ceiling file into a 0 MB cap', () => {
+    // A sandbox that stubs /sys/fs/cgroup with empty files: `Number('')` is 0, so
+    // dropping the empty-string term ships a 0 MB ceiling and labels the report
+    // cgroup-capped — a killer named off a file that measured nothing.
+    setSystemMemoryInfoReaderForTest(() => NO_HOST_PRESSURE)
+    setLinuxMemoryPressureStallReaderForTest(() => undefined)
+    fakeLinuxPseudoFiles({
+      '/proc/self/cgroup': SANDBOX_CGROUP_PATH,
+      '/sys/fs/cgroup/user.slice/user-1000.slice/app.slice/orca.scope/memory.current': '4200000000',
+      '/sys/fs/cgroup/user.slice/user-1000.slice/app.slice/orca.scope/memory.max': '',
+      '/sys/fs/cgroup/user.slice/user-1000.slice/app.slice/orca.scope/memory.high': ''
+    })
+
+    expect(readLinuxCgroupMemoryLimit('linux')).toEqual({
+      maxBytes: undefined,
+      highBytes: undefined,
+      currentBytes: 4_200_000_000,
+      oomKillCount: undefined,
+      maxEventCount: undefined,
+      highEventCount: undefined
+    })
+
+    const details = getSystemMemoryDetails('linux')
+
+    expect(details.systemMemoryCgroupMaxMB).toBeUndefined()
+    expect(details.systemMemoryCgroupHighMB).toBeUndefined()
+    expect(details.systemMemoryCgroupCurrentMB).toBe(4_005)
+    expect(details.systemMemoryPressureSignal).toBe('mem-available')
+  })
+
   it('says nothing rather than a row of undefineds when the files are garbage', () => {
     fakeLinuxPseudoFiles({
       '/proc/self/cgroup': SANDBOX_CGROUP_PATH,